i

Fundamental Studies of Bitumen Aeration

Juan Ma

Dissertation submitted to the faculty of the

Virginia Polytechnic Institute and State University

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

in

Mining Engineering

Roe-Hoan Yoon, Chair

Gregory T. Adel

Gerald H. Luttrell

William A. Ducker

Alan R. Esker

February 9, 2015

Blacksburg, Virginia

Keywords: wetting film, bitumen, air bubble, thinning, disjoining pressure

ii

Fundamental Studies of Bitumen Aeration

Juan Ma

ABSTRACT

In the oil sand industry, bitumen is separated from sands by aerating the heavy oil so that

it can float out of a flotation vessel, leaving the unaerated sands behind. A bubble-against-plate

apparatus equipped with a high-speed camera has been developed to record the optical interference

patterns of the wetting films formed on a flat surface and subsequently obtain the temporal and

spatial profiles of the films offline using the Reynolds lubrication theory. The technique has been

used to study the interaction mechanisms between air bubbles and bitumen. It has been found that

the film thinning kinetics increases in the order of asphaltene, bitumen, and maltene, and that the

kinetics increases sharply with increasing temperature.

In addition to obtaining kinetic information, the temporal and spatial profiles of the wetting

films have been used to derive appropriate hydrodynamic information that can be used to

determine the disjoining pressures () in the wetting films. The results obtained in the present

work show that < 0 for maltene and bitumen, while > 0 for asphaltene at temperatures in the

range of 22 to 80 °C. The disjoining pressure data have been analyzed by considering the

contributions from the hydrophobic and steric forces in addition to the classical DLVO forces. It

has been found that the hydrophobic force increases with increasing temperature, which

corroborates well with contact angle data. Dynamic contact angle measurements show that air

bubbles attach on bitumen with relatively small contact angles initially but increase sharply to

>90o. The extent and the kinetics of contact angle change increase sharply with increasing

temperature. These findings suggest that the primary role of temperature may be to increase

iii

bitumen hydrophobicity and hence hydrophobic force, which is the driving force for bubble-

bitumen interaction. A thermodynamic analysis carried out on the basis of the Frumkin-Derjaguin

isotherm suggests that the disjoining pressure will remain positive (and hence no flotation) until

the hydrophobic force becomes strong enough (due to temperature increase) to overcome the

positive disjoining pressure created during the course of bitumen liberation.

iv

Dedication

To Hongbo and Xuekai, for their love and support.

v

Acknowledgements

My most sincere gratitude goes to my dear advisor, Dr. Roe-Hoan Yoon, for his support

and guidance throughout the course of my Ph.D. program; for his inspiration, suggestion, and

encouragement in my research; and for sharing his philosophy in life.

Sincere thanks are also due to Dr. Gerald H. Luttrell, Dr. Gregory T. Adel, Dr. William A.

Ducker, and Dr. Alan R. Esker for their valuable suggestions and guidance throughout the entire

program.

Special acknowledgements go to Mr. Peter Cleyle and his associates at Suncor Energy,

Inc., Alberta, Canada, who provided bitumen samples and technical guidance during the course of

the project; and Prof. Zhenghe Xu and his colleagues at the University of Alberta for providing

useful information, including sample preparation methods.

The financial support from the CONRAD Bitumen Production Fundamental Research

Group in Canada is deeply appreciated.

I would like to extend my gratitude to colleagues and staffs of the Center for Advanced

Separation Technologies (CAST), especially to Drs. Lei Pan and Zuoli Li for their advice,

cooperation and training on how to use the equipment in our lab.

Finally, I would like to express my eternal gratitude to my parents, husband and son for

their love and support.

vi

Table of Contents

Chapter 1 Introduction................................................................................................................. 1

1.1 Background ......................................................................................................................... 1

1.2 Process Description ............................................................................................................. 2

1.3 Bitumen Composition ......................................................................................................... 6

1.3.1 Maltenes ......................................................................................................................... 7

1.3.1.1 Saturates .................................................................................................................. 7

1.3.1.2 Aromatics ................................................................................................................ 7

1.3.1.3 Resins ...................................................................................................................... 8

1.3.2 Asphaltenes .................................................................................................................... 8

1.4 Thin Liquid Film ................................................................................................................. 9

1.4.1 Disjoining Pressure ..................................................................................................... 10

1.4.2 Disjoining Pressure Isotherm ...................................................................................... 11

1.4.2.1 Van der Waals Force ............................................................................................. 12

1.4.2.2 Electrostatic Force ................................................................................................ 15

1.4.2.3 Steric Force ........................................................................................................... 19

1.4.2.4 Hydrophobic Force ............................................................................................... 21

1.5 Dissertation Outline .......................................................................................................... 23

References ................................................................................................................................ 25

Chapter 2 Temperature Effect on the Stability of Wetting Films of Water on Bitumen ..... 35

Abstract .................................................................................................................................... 35

2.1 Introduction ....................................................................................................................... 35

2.2 Experimental ..................................................................................................................... 39

2.2.1 Materials ...................................................................................................................... 39

2.2.2 Separation of Asphaltene and Maltene ........................................................................ 40

2.2.3 Preparation of Bitumen-, Maltene-, and Asphaltene-coated Surfaces ........................ 40

2.2.4 Measurement of Film Thinning Kinetics...................................................................... 41

2.2.5 Contact Angle Measurements on Bitumen Film .......................................................... 43

2.3 Results and Discussion ...................................................................................................... 44

2.3.1 Temperature Effect on the Film Thinning Kinetics...................................................... 44

2.3.2 Temperature Effet on Disjoning Pressure ................................................................... 49

2.3.3 Contact Angle Measurements at Different Temperatures ........................................... 49

vii

2.3.4 Thermodynamic Analysis Based on Acid-base Theory ................................................ 55

2.3.5 A Bitumen Aeration Model ........................................................................................... 57

2.4 Summary and Conclusion ................................................................................................ 62

References ................................................................................................................................ 63

Chapter 3 Colloidal Forces in Bitumen Aeration .................................................................... 67

Abstract .................................................................................................................................... 67

3.1 Introduction ....................................................................................................................... 67

3.2 Experimental ..................................................................................................................... 71

3.2.1 Materials ...................................................................................................................... 71

3.2.2 Separation of Asphaltene and Maltene ........................................................................ 71

3.2.3 Preparation of Bitumen-, Maltene-, and Asphaltene-coated Surfaces ........................ 72

3.2.4 Measurement of Film Thinning Kinetics...................................................................... 73

3.3 Results and Discussion ...................................................................................................... 75

3.3.1 Film Thinning Kinetics and Disjoining Pressure ........................................................ 75

3.3.2 Extended DLVO Theory ............................................................................................... 80

3.3.3 Disjoining Pressure Isotherms ..................................................................................... 82

3.3.4 Disjoining Pressure in Liberation ............................................................................... 92

3.4 Summary ............................................................................................................................ 94

References ................................................................................................................................ 94

Chapter 4 The Effect of Solution Chemistry on the Stability of Wetting Films on Bitumen

..................................................................................................................................................... 102

Abstract .................................................................................................................................. 102

4.1 Introduction ..................................................................................................................... 102

4.2 Experimental ................................................................................................................... 105

4.2.1 Materials .................................................................................................................... 105

4.2.2 Preparation of Bitumen-coated Surfaces ................................................................... 105

4.2.3 Measurement of Film Thinning Kinetics.................................................................... 106

4.3 Results and Discussion .................................................................................................... 107

4.3.1 Effect of Immersion Time ........................................................................................... 107

4.3.2 Effect of Electrolyte KCl ............................................................................................ 109

4.3.3 Effect of Solution pH .................................................................................................. 110

4.3.4 Effect of Calcium Ions ................................................................................................ 111

4.3.5 Effect of Magnesium Ions ........................................................................................... 112

viii

4.3.6 Effect of Approaching Speed on Film Thinning ......................................................... 113

4.4 Summary .......................................................................................................................... 115

References .............................................................................................................................. 115

Chapter 5 Summary and Future Work .................................................................................. 120

5.1 Research Revisits ............................................................................................................ 120

5.2 Conclusions ...................................................................................................................... 121

5.3 Original Contribution ..................................................................................................... 122

5.4 Future Work .................................................................................................................... 123

5.4.1 Studies of the Film Thinning Kinetics Using a New Cell........................................... 123

5.4.2 Measurement of the Length of the Asphaltene Tails in an Aqueous Solution ............ 125

5.4.3 Measurement of Surface Tension of Maltene and Maltene/water Interfacial Tension at

Different Temperatures ....................................................................................................... 125

5.4.4 Thermodynamic Studies on Surface Tension Components of Bitumen and Its

Components......................................................................................................................... 126

5.4.5 Aeration Studies in Micron Size ................................................................................. 126

References .............................................................................................................................. 126

ix

List of Figures

Figure 1.1 Generalized scheme for oil sands processing using water-based extraction

processes… ................................................................................................................ 3

Figure 1.2 Separation of bitumen into its various components ................................................... 7

Figure 2.1 Schematics of the modified TFBC apparatus used for the study of wetting films. . 42

Figure 2.2 Changes of Newton rings with time due to the decrease in the thickness of the

wetting film of water formed on maltene (a), bitumen (b) and asphaltene (c) surface,

while the air bubble was approaching the substrate at 80 °C. ................................. 44

Figure 2.3 Spatial and temporal profiles of the wetting films of water formed on maltene,

bitumen, and asphaltene surface, respectively, at 80 °C. The arrows represent the

critical thicknesses where the films rupture. ............................................................ 45

Figure 2.4 Effect of temperature on the kinetics of film thinning on bitumen in water at natural

pH of about 5.7. ....................................................................................................... 46

Figure 2.5 Effect of temperature on the kinetics of film thinning on maltene. ......................... 48

Figure 2.6 Effect of temperature on the kinetics of film thinning on asphaltene ...................... 48

Figure 2.7 Effect of temperature on the disjoining pressure isotherms ((h)) in the thin films

of water formed on maltene (a), asphaltene (b), and bitumen (c). The data have been

collected at the center of films. ................................................................................ 50

Figure 2.8 Dynamic water contact angles measured on the thin bitumen films at different

temperatures. The thickness of the bitumen films were 11 nm. ............................ 51

Figure 2.9 The kinetics of bubble-bitumen interactions at 22, 35, and 80 oC. The thickness of

the film was about 0.5 mm.. ..................................................................................... 52

Figure 2.10 Dynamic contact angles on the thick bitumen films at different temperatures in

water. The measurement was difficult due to the liquid (oil) collecting around the

three-phase contact lines as shown in Figure 2.9. Therefore, the contact angles were

measured by extending the bubble surface along the curvatures. ............................ 52

Figure 2.11 A model for the attachment of air bubble to the surface of bitumen at a high

temperature: a) initially, an air bubble attaches to a bitumen surface coated by

asphaltene with a low contact angle; b) maltene diffuses into the asphaltene layer on

the surface and increases the hydrophobicity and hence the contact angle. This is

x

possible due to the low viscosity of maltene at higher temperatures. Eventually, the

bitumen layer saturated with maltene begins to cover the bubble surfaces. ............ 53

Figure 2.12 A dynamic bitumen aeration model: a) an air bubble attaches to an asphaltene-

coated bitumen surface, forming a low contact angle; b) maltene diffuses into the

asphaltene layer on the surface and increases the contact angle of the layer, and

subsequently covers bubble surface; c) when the air bubble is covered by the

maltene-saturated asphaltene, the bubble enters bitumen droplet, completing

aeration. Thermodynamically, maltene-saturated asphaltene (or bitumen) spreads on

air bubble better than maltene. Thus, asphaltene may act as a spreading agent for

maltene. .................................................................................................................... 58

Figure 3.1 Schematics of the modified TFBC apparatus used for the study of wetting films. . 73

Figure 3.2 Spatial and temporal profiles of the wetting films of water formed on maltene (a),

asphaltene (b), and bitumen (c) at 55 °C. The arrows represent the critical

thicknesses (hcr) where the films rupture. ................................................................ 75

Figure 3.3 Kinetics of film thinning on maltene, bitumen, and asphaltene at 55 °C. It decreases

in the order of maltene, bitumen and asphaltene. .................................................... 76

Figure 3.4 Changes in the disjoining pressure () of the wetting films of water formed on

maltene (a), asphaltene (b), and bitumen(c) as functions of time and radial distance

(r) of the film from the film center at 55 °C. ........................................................... 78

Figure 3.5 Changes in the disjoining pressure () at the center of the wetting films of water

formed on maltene, bitumen, and asphaltene at 55 °C. Thermodynamically, the film

can rupture when < 0. .......................................................................................... 79

Figure 3.6 The disjoining pressure isotherm ((h)) obtained at the center of the wetting film of

water formed on maltene at 55 °C. The triangles represent the experimental data

presented in Figure 3.5. The solid line represents a fit to the extended DLVO theory,

which includes contributions from: van der Waals force (d) with a Hamaker

constant of A132 = -1 x 10-20 J; electrostatic force (e) with surface potential of ψ1 =

- 65 mV for maltene, surface potential of ψ2 = - 26 mV for air bubble, and Debye

length of κ-1 = 50 nm; and hydrophobic force (h) with fitting parameters of C = -

0.43 mN/m and D = 50 nm. ..................................................................................... 83

Figure 3.7 The disjoining pressure isotherm ((h)) obtained at the center of the wetting film of

water formed on asphaltene at 55 °C. The triangles represent the experimental data

presented in Figure 3.5. The solid line represents a fit to the extended DLVO theory,

which includes contributions from: van der Waals force (d) with a Hamaker

constant of A132 = -1 x 10-20 J; electrostatic force (e) with surface potential of ψ1 =

- 68 mV for asphaltene, surface potential of ψ2 = - 26 mV for air bubble, and Debye

xi

length of κ-1 = 65 nm; and repulsive steric force (s) between asphaltene polymers

with a brush length of L = 63 nm and a grafting distance of s = 23nm. .................. 85

Figure 3.8 The disjoining pressure isotherm ((h)) obtained at the center of the wetting film of

water formed on bitumen at 55 °C. The triangles represent the experimental data.

The solid line represents a fit to the extended DLVO theory, which includes

contributions from: van der Waals force (d) with a Hamaker constant of A132 = - 1

x 10-20 J; electrostatic force (e) with surface potential of ψ1 = - 57 mV for bitumen,

surface potential of ψ2 = - 30 mV for air bubble, and Debye length of κ-1 = 45 nm;

repulsive steric force s) between bitumen polymers with a brush length of L = 32

nm and a grafting distance of s = 26 nm, and hydrophobic force (h) with fitting

parameters of C = -0.2 mN/m and D = 54 nm. ........................................................ 86

Figure 3.9 Hydrophobic disjoining pressure isotherms (h) on bitumen at different

temperatures in water of a natual pH of about 5.7. .................................................. 88

Figure 3.10 Steric disjoining pressure isotherms (s) on bitumen at different temperatures in

water of a natual pH of about 5.7. ............................................................................ 91

Figure 3.11 a) For liberation, it is necessary to create a positive disjoining pressure in the TLFs

of water between bitumen and sand grains, i.e., > 0; for air bubble-bitumen

attachment, it is necessary to create negative disjoining pressures in the TLFs, i.e.,

< 0 ................................................................................................................ ………92

Figure 4.1 Schematic representation of the bubble-against-plate (BAP) apparatus used to

monitor the film thinning kinetics of wetting films. .............................................. 106

Figure 4.2 Thinning kinetics of the wetting films formed on bitumen at different immersion

times in water at pH 8.4 to 7.5. .............................................................................. 108

Figure 4.3 Effect of KCl on the thinning kinetics of the wetting films formed on bitumen in

aqueous solution of pH 8.4 to 7.5 and at 22 °C. .................................................... 109

Figure 4.4 Effect of solution pH on the thinning kinetics of the wetting films formed on

bitumen in a 1 mM KCl solution. The short arrows indicate the critical rupture

thicknesses (hcr) of the wetting films. .................................................................... 110

Figure 4.5 Effect of Ca2+ ions on the thinning kinetics of the wetting films formed on bitumen

in a 1 mM KCl solution at pH 8.4 to 7.5................................................................ 111

Figure 4.6 Effect of Mg2+ ions on the thinning kinetics of the wetting films formed on bitumen

in a 1 mM KCl solution at pH 8.4 to 7.5................................................................ 112

Figure 4.7 Temporal and spatial profiles of the wetting films formed bitumen in a 1 mM KCl

solution at two different approach speeds: a) 0.2 and b) 4.6 m/s. ....................... 113

xii

Figure 4.8 Comparison of the thinning kinetics obtained for dimpled and non-dimpled wetting

films in a 1 mM KCl solution. Black triangles represent the minimum film

thickness of the dimpled film, i.e., at the rim; black circles represent the film

thicknesses at the center of the dimpled films; and red circles, the minimum film

thicknesses of the non-dimpled films, i.e., at the center. ....................................... 114

Figure 5.1 New experimental set-up designed for the study of the thinning of wetting film of

water on bitumen.................................................................................................... 124

xiii

List of Tables

Table 2.1 Dynamic Viscosity Data for Bitumen, Maltene, n-Dodecane, Hexacane and

Naphthalene. ............................................................................................................ 54

Table 3.1 Parameters used to fit the disjoining pressure of the wetting film of water formed on

bitumen at different temperatures ............................................................................ 87

1

Chapter 1

Introduction

1.1 Background

Canadian oil sands reserves are estimated to contain as much as 1.7 to 2.5 trillion barrels

of crude oil, of which about 170 billion barrels are economically recoverable using existing

technologies. These numbers make Canada the third-largest country in the world in proven oil

reserves, after Saudi Arabia and Venezuela. This huge oil sands resource would be sufficient to

meet Canada’s need for crude oil for 186 years at an oil consumption rate of 2.5 million barrels

per day. Canada’s oil sands deposits, covering 141,000 square kilometers (54,132 square miles),

are located in three main regions within the province of Alberta: Peace River, Athabasca (Fort

McMurray area) and Cold Lake. Among them, the Athabasca area, which is located in the northeast

of Alberta, has the world’s largest oil sands deposit.1

Oil sand, also known as tar sand, bituminous sand, or bitumen-impregnated sandstone/rock,

is a type of unconventional petroleum deposit that contains a carbonaceous material with a high

viscosity at reservoir temperature. This carbonaceous material is normally referred to as bitumen.

At room temperature, bitumen, like cold molasses, is so viscous and difficult to flow that it must

be treated to reduce its viscosity before it can be transported. In the upgraders and refineries,

bitumen is further treated to produce usable fuels such as gasoline, jet fuel, heating oil and diesel

fuel.

Formed 50 to100 million years ago and first discovered in 1715, Alberta’s oil sands occur

naturally as a mixture of sands, fines (- 44µm particles including silt, clay and other minerals),

water and bitumen. It typically contains 7 to 14 wt.% bitumen, 3 to 5 wt.% water, and 83 to 88

2

wt.% solids. In an oil sand matrix, a film of water surrounds the sand grains, which in turn is

surrounded by bitumen. A portion of the clays and fines are associated with the bitumen at the

bitumen-water interface.2

Two predominant methods currently exist to extract bitumen from oil sands: surface

mining followed by bitumen extraction and in-situ extraction. Surface mining, or open-pit mining,

is used when oil sands are close to the surface, for deposits at a depth of less than 75 meters.

Surface mining represents the most significant source for bitumen production and is currently the

main extraction method. The problem, however, is that only about 20% of the oil sands are shallow

enough to be recovered by surface mining. The other method, in-situ extraction, is used to recover

oil sands that are hundreds of meters underground. Approximately, 80% of oil sands deposits are

situated deeply underground and are only recoverable through in-situ technologies. Most in-situ

methods involves injecting steam into the oil sands deposits to heat and soften the bitumen so that

it can be pumped to the surface through wells. Steam-Assisted Gravity Drainage (SAGD) and

Cyclic Steam Stimulation (CSS) are the two main in-situ methods. The present work is related to

surface mining, which is detailed in the following sections.

1.2 Process Description

Numerous scientists and engineers have investigated economic and efficient ways to

recover bitumen from oil sands. Among them, Dr. Karl Clark was the first to develop a successful

extraction process in the 1920s.3 His process, which is referred to as the Clark Hot Water Process,

laid the foundation for today’s oil sands extraction technologies. Despite some modifications over

the ensuing decades, Clark’s hot water process is still used in commercial operations by many oil

sand operators.

3

Figure 1.1 shows a generic flow diagram of the bitumen recovery process.4 The basic

operations of bitumen production include mining, utilities, extraction, froth treatment, water

management (tailings treatment) and upgrading, all of which are interrelated. For example, mining

affects extraction and the extraction, in turn, affects upgrading. As shown in Figure 1.1, in

commercial bitumen recovery, oil sands are mined in open-pit mines using large shovels and then

carried by trucks to the crushers, where large clumps of oil sands are broken down into

transportable sizes. The oil sands are then treated with a warm recycle process water to transform

dry oil sands into a slurry in mixing boxes, stirred tanks, cyclo-feeders or rotary breakers. Initially,

this slurry was then fed to tumblers; more recently, the tumblers have been replaced by pipelines,

known as hydrotransport process, which helps reduce the cost of transporting ore from mine sites

to processing plants. Whether in tumblers or hydrotransport pipelines, the lumps of oil sands are

Figure 1.1 Generalized scheme for oil sands processing using water-based extraction

processes.4

4

then broken and dispersed in water. While sand grains coated with bitumen are suspended in water,

the bitumen is detached (or liberated), so that it can interact with air bubbles and become aerated.

Chemical additives such as caustic soda can be added during the slurry conditioning stage.

The oil sand slurry is transported to large gravity separation vessels, known as primary

separation vessels (PSV), in which the aerated bitumen float to the surface while unaerated sands

sink to the bottom. The aerated bitumen floating to the top of the PSV is subsequently skimmed

off from the slurry. Small amounts of bitumen (or middlings) not recovered via the PSV are

recovered in a bank of mechanical flotation cells or cyclo-separators.5

The bitumen froth product, normally containing 60% bitumen, 30% water and 10% solids,

is de-aerated, diluted with hydrocarbon solvents (usually naphtha) to reduce bitumen density and

viscosity, and subsequently separated from the water and solids using centrifuge, cyclones, and/or

lamella thickener in the froth treatment process. Naphtha is recovered so that it can be used again.

Later, the diluted bitumen containing 1.5 to 2.5% water and 0.4 to 0.6% solids is upgraded via a

combination of thermal cracking and catalytic hydrocracking processes to obtain synthetic crude

oils (SCO) and hydrocarbon gases. An alternative is to use a paraffinic solvent (mainly hexane) at

a sufficiently high diluent-to-bitumen ratio to precipitate asphaltenes, forming composite

aggregates that trap the water and solids in the diluted bitumen froth. In this way diluted bitumen

with less than 0.5% solids plus water can be obtained.6

Tailings from the extraction plant, which is a process byproduct composed of water, clay,

sand, and small amount of residual bitumen, are then pumped to the tailings pond where water is

recycled to be used in the extraction plant. At Sunsor and Syncrude (the two major oil sand

companies in Canada), gypsum is added to mature fine solids to consolidate the fines together with

coarse sands into a non-segregating mixture that is subsequently disposed of in a geotechnical

5

manner. At Albian (a relatively new oil sand company), cyclones are used, with the underflow

(coarse tailings) being directly pumped into a tailings pond, while the overflow (fine tailings)

pumped to thickeners where they are treated with flocculants, then thickened and finally

discharged into a tailings pond.

The industrial bitumen-extraction process described above entails two fundamental steps:

the liberation of bitumen from sand grains, and the aeration of liberated bitumen to air bubbles.

Both steps are very important for the success of bitumen flotation. The bitumen liberation and

subsequent stabilization against heterocoagulation between the released bitumen and sand or clay

particles are prerequisites for bitumen flotation. NaOH is added to the slurry to create a weakly

alkaline environment that favors the liberation of bitumen from oil sands. There is an optimum

amount of added NaOH that leads to maximum bitumen recovery.2 The role of added NaOH was

reported to ionize organic acids in bitumen to generate natural surfactants, which are surface active

species that facilitate bitumen liberation.7

Once the bitumen is liberated, bitumen aeration, which involves the attachment and

engulfment of liberated bitumen to air bubbles, is critical to bitumen flotation since bitumen has

almost the same density as water over the temperature ranges used in the extraction process. The

attachment/engulfment of bitumen to air bubbles decreases the apparent density of bitumen so that

it can float to the top of the separation cell and be collected.1 Bitumen aeration occurs in the oil

sand slurry conditioning stage, at which point liberated bitumen attaches to air bubbles entrained

in oil sands or extra air bubbles added to the slurry, and in flotation cells used to extract remaining

bitumen in the middlings from the primary separation cell. Research shows that the aeration of

bitumen can also enhance the effective release of bitumen from sand grains.8

6

Despite the importance of bitumen aeration, several fundamental mechanisms involved in

this process are still not fully understood. For example, what is the driving force for the attachment

and subsequent engulfment of bitumen onto air bubbles? Does attachment have the same driving

force as engulfment? Will the conditions favorable for bitumen liberation enhance bitumen

aeration? What role will each bitumen component play in bitumen aeration? To answer these

questions, the present work focuses on studies of bitumen aeration—mainly the thinning of the

wetting film of water formed between bitumen and air bubble. Additionally, the engulfment of

bitumen over air bubbles is described.

1.3 Bitumen Composition

The fact that so many chemicals are present in bitumen makes its chemistry complex. The

elemental composition of bitumen differs from the crude source. Typically, bitumen contains

carbon (80 to 88 wt.%) and hydrogen atoms (8 to 12 wt.%), affording a hydrogen-to-carbon (H/C)

molar ratio of around 1.5, which is intermediate between that of aromatic structures and saturated

alkanes.9 In addition, bitumen contains heteroatoms such as sulphur (0 to 9 wt.%), nitrogen (0 to

2 wt.%), oxygen (0 to 2 wt.%), as well as trace metals such as vanadium (up to 2000 ppm) and

nickel (up to 200 ppm). Sulfur, generally the most ubiquitous polar atom, appears in the form of

sulfides, thiols, and to a lesser extent, sulfoxides; oxygen exists in the form of ketones, phenols

and, to a lesser extent, carboxylic acids; and nitrogen in the form of pyrrolic and pyridinic

structures. Most metals appear in the form of complexes such as metalloporphyrins.

Constituents of bitumen include maltenes and asphaltenes. Asphaltenes are defined as the

insoluble part of bitumen in low molecular weight paraffins (i.e., heptane, pentane, and hexane),

but soluble in light aromatic hydrocarbons (i.e., toluene and benzene). Maltenes are defined as the

soluble part of bitumen in both types of solvents, which can be further separated into saturates,

7

aromatics, and resins. At present, the composition of bitumen is usually given in so-called SARA

fractions, which refers to Saturates, Aromatics, Resins and Asphaltenes, as shown in Figure 1.2.

1.3.1 Maltenes

1.3.1.1 Saturates

Saturates have a H/C ratio close to 2, usually accounting for 5 to 15 wt.% of the total

bitumen.10 They contain a few crystalline n-alkanes, with a number-average molecular weight of

about 600 g/mol. Fourier-Transform Infra-Red spectroscopy (FTIR) shows that different

branching structures and some long aliphatic chains are present in saturates. At room temperature,

they appear as a colorless or lightly colored liquid. Their density at 20 °C is around 0.9 g/cm3.

1.3.1.2 Aromatics

Aromatics, also referred to as naphthene aromatics, are the most abundant components of

bitumen; in fact, they account for 30 to 45 wt.% of the total bitumen. The carbon skeleton of an

aromatic is slightly aliphatic with lightly condensed aromatic rings. Aromatics have a number-

average molecular weight of about 800 g/mol, and at room temperature they form a yellow to red

liquid. They are more viscous than saturates at the same temperature. They have a density close

(inferior) to 1 g/cm3 at 20 °C.10

Figure 1.2 Separation of bitumen into its various components

Bitumen

Maltene

Saturates Aromatics Resins

Asphaltene

8

1.3.1.3 Resins

Resins also account for 30 to 45 wt.% of the total bitumen; moreover, depending on choice

of solvent they could even outnumber the percentage of aromatics. Resins have a H/C ratio

between 1.38 and 1.69, with a number-average molecular weight of about 1100 g/mol.10 It has

been shown that resins have a similar composition to that of asphaltene, but exhibit a less

condensed aromatic structure and a lower molar mass. They typically contain fused aromatic rings,

with a most probable structure corresponding to 2 to 4 fused rings, and a few polar groups. When

saturates and aromatics are oily liquids at room temperature, the resins form a black solid. Their

density at 20 °C is close to 1.07 g/cm3.

1.3.2 Asphaltenes

Asphaltenes represent a solubility class of macromolecules present in crude oil and

bitumen. They have a H/C ratio between 0.98 and 1.56 depending on the asphaltene source, and

usually gather traces of transition metals of the whole bitumen (e.g., Ni, Va, Fe), thereby forming

complexes such as metallo-porphyrins.9 They account for 5 to 20 wt.% of the total bitumen. The

molecular weight of asphaltenes has been controversial for decades, with estimates ranging from

700 to 3500 g/mol.11 Some estimates of the average molecular weight of asphaltene are as high as

9500 g/mol.12 Asphaltenes are by far the most studied bitumen fraction due to their viscosity-

building role and their importance in the processing of crude oil. They exist as a black powder at

room temperature and are largely responsible for the black color of bitumen. They have a density

of about 1.15 g/cm3 at 20 °C. Moreover, asphaltenes contain fused aromatic rings, most probably

4 to 10 fused rings, together with some pending aliphatic chains and polar groups such as carboxyl,

amide, thiol, and hydroxyl groups.9 In comparison to other bitumen molecules, asphaltenes contain

9

more condensed aromatic rings and more polar groups. Because of their many condensed aromatic

rings, asphaltenes form almost planar molecules.

1.4 Thin Liquid Film

A thin liquid film (TLF) is formed when two interfaces of a liquid are brought into close

proximity, normally a few hundred nanometers or less. TLFs are ubiquitous in industrial processes

and among scholarly researchers, and also play a vital role in our daily lives. There are six types

of TLF, which are categorized according to the phases of the interface:

1) Solid/liquid/solid film: A film confined between two solid surfaces, which is very common

in colloidal chemistry. For instance, a water film between two quartz or mica surfaces is

often studied as a model system by atomic force microscopy (AFM) to investigate surface

forces. Another example is the oil film between rocks in an oil reservoir, which is of interest

to the oil and gas industry.

2) Liquid/liquid/liquid film: A film between two liquid droplets that is immiscible with the

liquid in between. This type of TLF is common in emulsions, such as milk (an emulsion of

fat in water and proteins).

3) Gas/liquid/gas film: A foam lamella separating gas phases, which is frequently found in

soap, shampoo, and the froth flotation industry.

4) Solid/liquid/liquid film: A film between a solid surface and a liquid surface. For instance,

the water film between rock and oil in an oil reservoir.

5) Solid/liquid/gas film: A film between a solid and a gas bubble—i.e., a wetting film. For

example, the water film between a mineral particle and an air bubble in flotation.

6) Liquid/liquid/gas film: A film between a liquid surface and a gas bubble, such as the water

film between an oil and air bubble.

10

There are many different types of TLFs associated with the bitumen extraction process:

those between bitumen and air bubble, between bitumen and sand grain, between bitumen and clay

particle such as kaolinite, illite or montmorillonite, between air bubble and sand grain, between air

bubble and clay particle, between sand grain and clay particle, between two bitumen droplets,

between two air bubbles, between two sand grains, and between two similar or different clay

particles. In the present work, we focus on the wetting film between bitumen and air bubble. Other

types of TLFs will only be discussed briefly (or not included in this study).

1.4.1 Disjoining Pressure

TLFs are known to be unstable and seek to thicken or thin themselves depending on the

pressure in the film ( ) relative to that in the adjacent bulk ( ) of the same fluid under the

same thermodynamic conditions. Such a difference is defined, originally by Derjaguin in 1955, as

the “disjoining pressure” Π(h),13 h being the thickness of the TLF,

[1.1]

Thermodynamically, disjoining pressure can be defined as the change in excess Gibbs free

energy per unit area of a flat TLF (G) with the film thickness (h),14, 15

[1.2]

at constant pressure (P), temperature (T), and chemical potential of solutes (μs).

Disjoining pressure quantifies the driving force for spontaneous thinning or thickening of

a film. With a positive disjoining pressure, i.e., Π > 0, the film thickens; with a negative disjoining

pressure, i.e., Π < 0, it thins. When Π = 0, it is in an equilibrium state. The process of thickening

is equivalent to separating (or “disjoining”) its bounding surfaces, hence its name. When h

becomes sufficiently large that the liquid in the film behaves the same as in the bulk, Π tends to

be zero.

filmPbulkP

bulkfilm PPh )(

STPhGh ,,)/()(

11

Disjoining pressure determines the stability of a TLF. A TLF should be stable

thermodynamically when the disjoining pressure derivative is negative, i.e., ∂Π/∂h < 0. In other

words, the disjoining pressure vs. h curve should have a negative slope. When the disjoining

pressure derivative is positive, i.e., ∂Π/∂h > 0, a TLF should be unstable, thin by itself, and rupture

catastrophically, such as the wetting film between a hydrophobic particle and an air bubble in

flotation.

The most common method for measuring a film with a thinning thickness is to use optical

interference techniques in conjunction with microscopy, which was originally developed by

Derjaguin and Kusakov16 to measure > 0. In 1967, Laskowski and Kitchener17 recognized that

the disjoining pressure in wetting films must be negative for bubble-particle attachment to occur

in flotation. However, the authors recognized the difficulty of measuring negative disjoining

pressure due to several reasons—principally fast kinetics of film thinning, deformation of the

air/water interface, and complex interactions between hydrodynamic and surface forces. It was

only recently that CAST developed a technique for measuring negative disjoining pressure using

a modified thin film pressure balance (TFPB) technique with high-speed video microscopy.18, 19

This new method is based on monitoring the profiles of the wetting films in nanoscale, and then

analyzing the profiles using the Reynolds lubrication theory to determine various pressures. Such

a technique is possible because bubbles deform in response to changes in pressure.

1.4.2 Disjoining Pressure Isotherm

As indicated in Eqs. [1.1] and [1.2], disjoining pressure Π is a function of the film thickness

h. Π(h), called the disjoining pressure isotherm, depends on the makeup of the film, the adjoining

bulk phases and the interfaces between them.

12

There are many contributors to disjoining pressure. The classical Derjaguin-Landau-

Verwey-Overbeek (DLVO) theory is based on the contribution from the van der Waals force

present in a TLF ( d ) and the electrostatic force between overlapping electrical double layers

( e ), as shown below,

ed [1.3]

DLVO theory is used to explain the stability of a colloidal system, but only to a limited

degree. Specifically, discrepancies between DLVO theory and experimental results have been

reported in a number of cases,20, 21 indicating that other contributions must also be considered, as

shown in the following equation,

[1.4]

where , , , , represent the contribution from hydrophobic force, steric force,

hydration force, and hydrogen bonding, respectively. There might be other forces as well, such as

supramolecular structuring forces.

1.4.2.1 Van der Waals Force

Van der Waals dispersion forces are ever-present between two macroscopic objects and

can be important both for both small and large separations. There are three origins for Van der

Waals forces: 1) electrostatic interactions between permanent charge distributions in the molecules,

such as dipoles, quadrupoles, etc. (Keesom’s dipole – dipole interaction); 2) electrostatic

interactions between permanent charge distributions and induced charge distributions (Debye’s

dipole – induced dipole interaction); and 3) the dispersion forces associated with the oscillations

of the orbital electrons and interactions with the synchronous induced dipoles in neighboring

molecules (London’s fluctuating dipole-induced dipole interaction ).22

Hhydrationshed

h shydration

H

13

Hamaker,23 who was responsible for elucidating the mechanisms of the van der Waals

interaction, suggested that the van der Waals force between two macroscopic objects could be

obtained by adding up the forces acting between all the molecules in one body and all those in the

other body. The Hamaker constant, A, which is often used in interaction laws, is conventionally

defined as

[1.5]

where and are the number of atoms per unit volume (number density) in the two bodies,

and C is the coefficient in the atom-atom pair potential. Typical values for the Hamaker constants

of condensed phases, either solid or liquid, are about 10-19 or 10-20 J for interactions in a vacuum

or air. For gas phases, the Hamaker constant is generally taken to be zero due to the low molecular

density in such media.

The pairwise additivity assumptions in Eq. [1.5] ignore the influence of neighboring atoms

on the interaction between any pair of atoms. The Lifshitz theory24 avoided the additivity problem

and provided an alternative way to calculate the nonretarded Hamaker constant macroscopically

for the interaction of material (1) and (2) in the medium (3), as shown below

[1.6]

where k is Boltzmann constant, T is the absolute temperature (in Kelvin), is the Planck’s

constant (6.63×10-34J·s), is main electronic adsorption frequency (3×1015 Hz), , and

represent the refractive indices of the three media, and and refer to the static dielectric

constants of the three media. Eq. [1.6] is an approximate expression based on the assumption that

the absorption frequencies of all three media are the same.

21

2 CA

1 2

2/12

3

2

2

2/12

3

2

1

2/12

3

2

2

2/12

3

2

1

2

3

2

2

2

3

2

1

32

32

31

31132

)()()()(

))((

28

3))((

4

3

nnnnnnnn

nnnnhkTA

ep

ph

e 1n 2n 3n

1 2 3

14

For the symmetric case of two surfaces composed of the same material (1) in a medium

(3), one obtains

[1.7]

Hamaker constant is often used to calculate the disjoining pressure in a TLF,

[1.8]

It is useful to consider the general situation of two different materials (1) and (2) in a liquid

film (3). The former may be solids, liquids immiscible with (3) or gases, or any combination

thereof. The Hamaker constant of the film for computation of the disjoining pressure isotherm is25

[1.9]

The first two terms consider the interactions of the adjoining phases with each other and the film

molecules with themselves, while the last two consider the interaction of the film with its adjoining

phases.

When dispersion forces dominate the interactions between two materials, the geometric

mean combining rule is applicable, as shown below

[1.10]

where i and j mean different materials. It should be noted that when the medium between two

materials is with high dielectric constants, the use of combining rules is not recommended.

Substituting Eq. [1.10] to Eq. [1.9], one can obtain

[1.11]

From Eq. [1.11], one can easily find that for different materials, the Hamaker constant

could be either positive or negative. According to Eq. [1.8], positive values suggest van der Waals

2/32

3

2

1

22

3

2

12

31

31131

)(

)(

216

3)(

4

3

nn

nnhkTA

ep

3

132

6 h

Ad

23133312132 AAAAA

jjiiij AAA

))(( 33223311132 AAAAA

15

attraction, while negative values suggest van der Waals repulsion. If A33 is intermediate in value

between A11 and A22, the Hamaker constant will be negative, which is the case for flotation.

If the materials separating the liquid film are the same,

[1.12]

A131 is thus always positive in spite of the relative values of A11 and A33. Therefore, van der

Waals interactions are always attractive for like materials, such as foam films and films between

two like particles or liquid droplets. The greater the difference between Hamaker constants of the

material and the medium, the greater the van der Waals attraction.

In bitumen flotation, where the case is bubble (1)-bitumen (2) interaction in the medium of

water (3), the Hamaker constant of air bubble tends to zero (A11 = 0), as mentioned earlier, and the

Hamaker constant of bitumen is greater than that of water (A22 >A33). Therefore, the Hamaker

constant of the film is negative (A132 <0) according to Eq. [1.11], indicating repulsive van der

Waals interactions between air bubbles and bitumen in flotation.

1.4.2.2 Electrostatic Force

Interfaces rarely carry a net charge. They are normally charged because of following

mechanisms: 1) preferential adsorption/desorption of lattice ions, 2) specific adsorption of ions, 3)

ionization of surface functional groups, 4) isomorphic substitution, and 5) depletion of electrons.

When a surface acquires a charge, counterions gather in the vicinity of the charged surface and

neutralize the charge, forming an “electric double layer.” If two surfaces bearing an electric double

layer approach one another, their overlap can contribute to the total disjoining pressure.

The electrical double-layer interaction between a particle and an air bubble can be

determined from the Poisson-Boltzmann equation, as shown below:

2

3311131 )( AAA

16

[1.13]

where is the dielectric constant of the solution, is the permittivity of vacuum, is the

potential, e is the charge on electron, is the number per unit volume of the electrolyte ions

of type i with valence . Two boundary conditions are needed to solve the Poisson-Boltzmann

equation.

The electrical double-layer interaction also depends on the charging mechanisms at the

surfaces. Three cases are considered:

1) Both surfaces at constant surface potentials.

In this case, the surface potentials are known and remain the same for all separations. Zeta

potential is often used rather than surface potential. The disjoining pressure between a particle and

an air bubble can be calculated using the Hogg-Healey-Fuerstenau (HHF) approximation:26, 27

[1.14]

where is the reciprocal Debye length, and are the zeta potentials at the solid/water and

air/water interfaces, respectively. Eq. [1.14] holds exactly for surface potentials less than 25/zi mV

(the condition of the Debye-Hückel linearization) and for small-scaled separations (the condition

of the Derjaguin approximation). A comparison between the HHF expression and the exact

solution of the Poisson-Boltzmann equation shows that even at 75 to 100 mV the divergence is not

excessive except for very short distances between the surfaces.26

For two surfaces with potentials of the unlike sign, the interaction is attractive to each other

at all separation distances, as expected. For two surfaces with potentials of the same sign but of

unequal magnitude, the interaction is repulsive at large separations, but attractive at small

kT

eznze i

ii

exp)(2

0

0

)(in

iz

)coth(2)(cos)()sinh(2

21

2

2

2

1

2

0 hhechh

e

1 2

17

separations. When the surfaces have potentials of the same sign and of equal magnitude, the

interaction is repulsive at all separations.27

2) Both surfaces at constant surface charges.

If a system cannot regulate its charge during the interaction, a constant charge interaction

is appropriate. In this case, charge per unit area on the surfaces (charge densities) remains constant.

The disjoining pressure is given by28

[1.15]

Where and are the zeta potentials at the isolated solid/water and air/water interfaces

before the interaction, respectively.

When the two surfaces have unlike sign charges of unequal magnitude, the interaction is

attractive at large separations, but becomes repulsive at small separations. When the two surfaces

are of unlike sign charges with equal magnitude, the interaction is attractive at all separations.

When the two surfaces have the same sign charges, the interaction is repulsive at all separation

distances.29

3) One surface at constant potential and the other at constant charge.

In this case, one surface has constant potential at all separations while the other has constant

surface charge. The disjoining pressure can be obtained by29

[1.16]

The electrical double-layer interaction in this case falls between that at constant potential and at

constant charge. As shown above, for two surfaces having the same initial potential, the

electrostatic interactions are different for systems due to the differences in the charging

mechanisms of their surfaces.

)(sinh

)cosh(2

2 2

2

2

2

121

2

0

h

he

1 2

)(cosh

)()sinh(2

2 2

2

2

2

121

2

0

h

he

18

Given the fact that the third case is complicated, the first and second cases have been

favored in scholarly research. The constant potential case (#1) represents an ideal regulation of the

surface charge. When the surface charge cannot be regulated during the interaction, a constant

charge model should be used. In flotation, if a particle slides over an air bubble, the location of the

particle-bubble interaction sites changes continuously, making it difficult to achieve a perfect

regulation of the surface charge. In such cases the interaction at constant surface charge is more

appropriate than at constant surface potential. If the location of the particle-bubble interaction does

not change over the bubble surface, constant surface potential interaction may occur.

According to Eqs. [1.14] to [1.16], one must know surface potential to calculate the

electrostatic double-layer interaction. Surface potential is often substituted by zeta potential, which

can be measured by electrokinetic methods such as electrophoresis, electroosmosis, or streaming

potential. In bitumen flotation, the zeta potential of bitumen emulsion is mainly measured using

electrokinetic methods. Research shows that the bitumen zeta potential becomes increasingly more

negative when the solution pH increases from 4 to 6, but then levels off at pH>8.30, 31 The iso-

electric point of bitumen in water has been reported to be about pH = 3.

Measuring the zeta potential of gas bubbles is more difficult than that of solid particles due

to the inconveniencies associated with the generation of gas bubbles in the measuring cells and the

rise of bubbles in liquid solutions. Different values have been reported depending on the solution

pH, electrolyte concentration, surfactant addition, etc.32-37 In general, gas bubbles are negatively

charged in pure water and inorganic solutions such as NaCl and KCl at pH > 2-3. The magnitude

of the zeta potential decreases with increasing salt concentration, suggesting that Na+, K+, Cl- ions

act as indifferent ions that control the effective diffuse layer thickness, but do not specifically

adsorb on the gas-water interface. When surfactants are present in the solution, the zeta potential

19

of gas bubbles is determined by the type of surfactant—specifically, negative for an anionic

surfactant and positive for a cationic surfactant.

1.4.2.3 Steric Force

Although DLVO theory can be applied to many systems, the situation becomes more

complicated when considering flotation. In this case, non-DLVO forces such as steric force and

hydrophobic force must also be considered. When bitumen is in contact with water, asphaltene is

extracted from bitumen and deposited on the bitumen/water interface. The polar groups of

asphaltene are exposed toward the aqueous phase.38 In this manner, the bitumen surface would

become relaxed or swelled, forming brush-like structures. It has been reported that asphaltene is

responsible for the steric repulsion in bitumen flotation systems.38-41

The steric interaction has been studied both theoretically and experimentally.42-44

Normally, stabilizing polymers contain anchor groups and buoy groups. Anchor groups are

components that are attached either chemically or physically to a surface to prevent the escape of

polymers as polymer-coated surfaces approach each other. Buoy groups, or stabilizing moieties,

are the adsorbed layers that are as solvent-compatible as possible and have sufficient molar mass

to provide needed adlayer thickness. Anchor groups usually account 10 to 25% of the total

adsorbing macromolecule, with the remaining volume occupied by the buoy groups.

When two polymer-coated surfaces are brought into contact and start to interpenetrate, the

potential energy of the steric interaction may be comprised of: 1) mixing effect — the excess

chemical potential of the adsorbate and solvent due to overlapping of layers; 2) volume restriction

effect — elastic energy arising from compression of the polymer layers, which makes it difficult

to press two polymer layers together; or 3) changes in surface free energy, which could occur from

moving polymers laterally away from the surface or because of the desorption of polymers.45 The

20

third factor, however, can be discarded since typically the adsorbed molecules are assumed to be

anchored irreversibly and locally.

The most important properties of the system in terms of steric repulsion are 1) the adlayer

thickness or the polymer tail length L, and 2) the solvency of the buoy blocks of the polymer in

the medium. These two properties are related to each other. While the adlayer thickness depends

primarily on the molar mass of the buoy groups of the polymer, solvency can also play a role. In a

good solvent, interactions between buoy groups of polymers and solvent molecules are

energetically favorable; thus, the polymer coils will expand or extend. In a poor solvent, polymers

prefer to interact with themselves and the polymer coils will shrink. The length of polymer tails is

roughly equivalent to the end-to-end distance of these free polymer segments under the same

conditions as the system. At low surface coverage and in a good solvent, the thickness of the

adsorbed polymer layer should be roughly equal to the Flory radius (RF):

[1.17]

where l and n are the length and the number of the monomers, respectively.

At higher coverage, the grafted chains are so close to each other that they are forced to

extend away from the surface much farther than RF, thereby forming a “brush layer.” For a brush

in a good solvent, L becomes:46

[1.18]

where s is the mean distance between attachment points.

Once two brush-bearing surfaces are within a distance of 2L from each other, there will be

a repulsive pressure between them. According to the Alexander-de Gennes theory,44 the disjoining

pressure is given by:

lnRF

5/3

3/5

s

RsL F

21

4/39/4

3)

2()

2(

L

h

h

L

s

kTs

[1.19]

1.4.2.4 Hydrophobic Force

In addition to steric force, another non-DLVO force—namely, the hydrophobic force—is

also critical for bitumen flotation because both bitumen and air bubble surfaces are hydrophobic.

Israelachvili and Pashley47 were the first to directly measure the short-range attractive hydrophobic

force between two curved mica surfaces using surface force apparatus (SFA). The researchers

rendered the mica surfaces moderately hydrophobic using a cationic surfactant

hexadecyltrimethylammonium bromide (CTAB), having an advancing water contact angle of

about 60°.48 Since then, many other investigators have measured both short- and long-range

hydrophobic forces between two solid surfaces using SFA and atomic force microscopy (AFM).49-

52 All results indicate the presence of an attractive hydrophobic force between hydrophobic solid

surfaces.

Measuring the surface forces between a solid surface and an air bubble were made possible

by AFM.53 Assessing particle-bubble interactions using AFM was first reported by Ducker et al.54

and Butt55 in 1994; other related studies soon followed.56-62 Attractive forces were observed

between a hydrophobized silica particle and an air bubble in water.

The hydrophobic forces ( ) measured in experiments are commonly represented using

the following form,47

[1.20]

where R is the radius of curvature of the hydrophobic surfaces interacting with each other, C and

D (decay length) are fitting parameters.

According to the Derjaguin approximation,63 which is shown in the following equation,

hF

)exp(D

hC

R

Fh

22

[1.21]

one can obtain Eq. [1.22] to calculate the disjoining pressure contributed by hydrophobic force:

[1.22]

The origin of the hydrophobic force is still controversial. Some believe it is of entropic

origin, arising from the rearrangement of water molecules at hydrophobic surfaces.64-66 When

water molecules come into contact with a small nonpolar molecule or bubble, they re-orientate or

re-structure to form “clathrate cages” or “gas hydrates.”67 When water molecules are in contact

with a hydrophobic-water interface, including the air-water interface, they also form clathrate-like

structures.68, 69 When two hydrophobic surfaces approach each other, the water structure changes

in the overlapping boundary layers of water.70 Based on this concept, Eriksson et al.66 derived a

theoretical model for hydrophobic force, which assumes that the intervening water between two

surfaces is progressively more ordered with decreasing h. Most recently, Yoon and coworkers71-73

determined the thermodynamic functions of hydrophobic interactions by conducting surface force

measurements at several different temperatures. The researchers showed that macroscopic

hydrophobic interactions entail decreases in both the excess entropy (Sf) and the excess enthalpy

(Hf) of thin liquid films (TLF) of water confined between hydrophobic surfaces.

Other investigators believe that hydrophobic force is due to the bridging of nanoscopic

bubbles preexisting on the hydrophobic force. This theory is supported by three important

observations: 1) discontinuities (or “steps”) on force vs. distance curves,52, 74 2) the disappearance

of attractive forces in deaerated water,75 and 3) the presence of nanobubbles on hydrophobic

surfaces detected by AFM imaging.76 In contrast, other researchers argue that 1) steps on the force

curve can be avoided by modifying the conditions and procedures involved in force

h

hh dh

R

F2

)exp(2 D

h

D

Ch

23

measurement,77 2) long-range attractions still exist in degassed solutions,78, 79 and 3) AFM studies

do not show nanobubbles on hydrophobic surfaces;80 rather, optical measurements confirm that

water between two hydrophobic surfaces has the same refractive index as the bulk water.81

There are other possibilities for the origin of hydrophobic force, including electrostatic

origin,82 separation-induced cavitation,83 hydrodynamic fluctuating correlation,84 etc.

In bitumen flotation, all , and are positive (or repulsive); therefore, the total

disjoining pressure does not afford a negative value that is required for bitumen-bubble attachment.

Adding a hydrophobic component to the disjoining pressure allows the total disjoining pressure to

be negative when is negative.

1.5 Dissertation Outline

Despite the importance of bitumen aeration in water-based bitumen extraction processes,

some of the fundamental mechanisms involved in bitumen aeration are still not fully understood.

Thus, the principal objective of this thesis work is to investigate the thinning of wetting films

formed between bitumen and air bubbles. To facilitate this objective, we developed a bubble-

against-plate apparatus equipped with a high-speed camera to record the optical interference

patterns of the wetting films formed on a flat bitumen surface. It is necessary to use a high-speed

camera as the wetting films formed on hydrophobic surfaces are unstable and hence thin fast. The

recorded interference patterns were then used to obtain the temporal and spatial profiles of the

films offline. The film profiles were then analyzed to determine the disjoining pressure () using

the method and algorithms developed by Pan, et al.18 These authors derived the following

expression for disjoining pressure,

d e s

h

24

r

r

r

rdrdr

t

hr

rhr

hr

rrR 03

112

2

[1.23]

where is the air/water interfacial tension, R the bubble radius, r the radial position, h the wetting

film thickness, and µ the fluid viscosity.

Chapter 1 gives an overview of the state-of-the-art oil sands extraction process and a brief

review of the recent scientific research conducted on the role of colloidal forces in the extraction

processes.

Chapter 2 describes the temperature effect on the kinetics of the thinning of wetting films

of water on bitumen. A modified thin film pressure balance (TFPB) technique has been used to

monitor the changes in the profiles of the TLFs between an air bubble and a flat substrate coated

by bitumen, maltene, or asphaltene at temperatures ranging from 22 to 80 °C. The kinetic

information in the vertical and radial directions, i.e., 𝜕ℎ 𝜕𝑡⁄ and 𝜕ℎ 𝜕𝑟⁄ , respectively, have been

derived from the film profiles to determine the disjoining pressure () in the TLFs. In addition,

water contact angles on bitumen surfaces at temperatures in the range of 22 to 80 °C have been

measured using captive bubble method. Based on these results and the thermodynamic data

available in the literature, a model for bitumen aeration is proposed.

Chapter 3 studies the role of colloidal forces in bitumen aeration. The thin film pressure

balance (TFPB) technique has been used i) to study the kinetics of thinning of the wetting films

formed on bitumen and its hydrophilic and hydrophobic components, i.e., asphaltene, and maltene,

and ii) to determine the disjoining pressures () in the films from the curvature changes recorded

during the process of film thinning. The results will be analyzed in view of the extended DLVO

theory, which includes the contributions from the electrical double-layer force, van der Waals force,

hydrophobic force, and steric force that may be present in the TLFs.

25

Chapter 4 deals with the effects of solution chemistry on the stability of wetting films on

bitumen. A new bubble-against-plate apparatus has been developed and used to study effects of

immersion time, pH, KCl, Ca2+, Mg2+ ion concentrations, and the approach speed on bubble-

surface interactions.

Chapter 5 includes the summary of the present work and suggestions for future work.

References

1. Masliyah, J.H., Z. Xu, and J.A. Czarnecki, Handbook on Theory and Practice of Bitumen

Recovery from Athabasca Oil Sands, Kingsley Publishing Services, 2011.

2. Sanford, E.C., Processibility of athabasca oil sand: Interrelationship between oil sand

fine solids, process aids, mechanical energy and oil sand age after mining, The Canadian

Journal of Chemical Engineering. 61(4), p. 554-567, 1983.

3. Clark, K. and D. Pasternack, Hot Water Seperation of Bitumen from Alberta Bituminous

Sand, Industrial & Engineering Chemistry. 24(12), p. 1410-1416, 1932.

4. Masliyah, J., et al., Understanding Water‐Based Bitumen Extraction from Athabasca

Oil Sands, The Canadian Journal of Chemical Engineering. 82(4), p. 628-654, 2004.

5. Cymerman, G.J. and T. Kwong, Improvements in the Oil Recovery Flotation Process at

Syncrude Canada Ltd., in Processing of Hydrophobic Minerals and Fine Coals:

Proceedings of the 1st UBSMcGill Bi-Annual International Symlposium on

Fundamentals of Mineral Processing, CIM, Vancouver, BC, 1995.

6. Tipman, R.N. and Y.-C. Long, patent US 5876492, 1999.

7. Sanford, E.C. and S. F.A., Processability of Athabasca Tar Sand Using a Batch

Extraction Unit: The Role of NaOH, CIM Bull. 72(803), p. 164-169, 1979.

26

8. Drelich, J., et al., The role of gas bubbles in bitumen release during oil sand digestion,

Fuel. 74(8), p. 1150-1155, 1995.

9. Lesueur, D., The colloidal structure of bitumen: Consequences on the rheology and on

the mechanisms of bitumen modification, Advances in colloid and Interface Science.

145(1), p. 42-82, 2009.

10. Corbett, L.W., Composition of asphalt based on generic fractionation, using solvent

deasphaltening, elution-adsorption chromatography, and densimetric characterization,

Analytical Chemistry. 41(4), p. 576-579, 1969.

11. Peramanu, S., B.B. Pruden, and P. Rahimi, Molecular Weight and Specific Gravity

Distributions for Athabasca and Cold Lake Bitumens and Their Saturate, Aromatic,

Resin, and Asphaltene Fractions, Industrial & Engineering Chemistry Research. 38(8), p.

3121-3130, 1999.

12. Leon, O., et al., Electrophoretic mobility and stabilization of asphaltenes in low

conductivity media, Petroleum science and technology. 18(7-8), p. 913-927, 2000.

13. Derjaguin, B.V., Kolloid Zh. 17(205), 1955.

14. Derjaguin, B.V., Theory of stability of colloids and thin films, Consultants Bureau New

York, 1989.

15. Eriksson, J.C. and R.H. Yoon, Role of Surface Forces - Part I, Colloid and Interface

Science Series, T.F. Tadros, Editor, p. 99, Wiley-VCH, 2007.

16. Derjaguin, B.V. and M.M. Kusakov, An experimental investigation of polymolecular

solvate (adsorbed) films as applied to the development of a mathematical theory of

stability of colloids, Acta Physicochim. URSS. 10 p. 25, 1939.

27

17. Laskowski, J. and J. Kitchener, The hydrophilic—hydrophobic transition on silica, J. of

Colloid and Interface Sci. 29(4), p. 670-679, 1969.

18. Pan, L., S. Jung, and R.H. Yoon, Effect of hydrophobicity on the stability of the wetting

films of water formed on gold surfaces, Journal of colloid and interface science. 361(1),

p. 321-330, 2011.

19. Pan, L., S. Jung, and R.H. Yoon, A fundamental study on the role of collector in the

kinetics of bubble–particle interaction, International Journal of Mineral Processing. 106

p. 37-41, 2012.

20. Tchaliovska, S., et al., Interactions in Equilibrium Free Films of Aqueous

Dodecylammonium Chloride Solutions, Journal of Colloid and Interface Science. 168(1),

p. 190-197, 1994.

21. Lyklema, J. and K.J. Mysels, A Study of Double Layer Repulsion and van der Waals

Attraction in Soap Films, Journal of the American Chemical Society. 87(12), p. 2539-

2546, 1965.

22. Berg, J.C., An introduction to interfaces & colloids: the bridge to nanoscience, World

Scientific, 2010.

23. Hamaker, H., The London—van der Waals attraction between spherical particles,

physica. 4(10), p. 1058-1072, 1937.

24. Lifshitz, E.M., J. Exp. Theoret. Phys. (USSR). 29(94), 1955.

25. Israelachvili, J.N., Intermolecular and Surface Forces, 2nd ed, Academic press, London,

1992.

26. Hogg, R., T.W. Healy, and D.W. Fuerstenau, Mutual coagulation of colloidal

dispersions, Transactions of the Faraday Society. 62(0), p. 1638-1651, 1966.

28

27. Nguyen, A. and H.J. Schulze, Colloidal science of flotation, p. 332, CRC Press, 2003.

28. Nguyen, A. and H.J. Schulze, Colloidal science of flotation, p. 334, CRC Press, 2003.

29. Nguyen, A. and H.J. Schulze, Colloidal science of flotation, p. 337, CRC Press, 2003.

30. Liu, J., et al., Bitumen–clay interactions in aqueous media studied by zeta potential

distribution measurement, Journal of colloid and interface science. 252(2), p. 409-418,

2002.

31. Hall, A., S. Collins, and J. Melrose, Stability of aqueous wetting films in Athabasca tar

sands, Society of Petroleum Engineers Journal. 23(2), p. 249-258, 1983.

32. McShea, J.A. and I. Callaghan, Electrokinetic potentials at the gas-aqueous interface by

spinning cylinder electrophoresis, Colloid & Polymer Science. 261(9), p. 757-766, 1983.

33. Yoon, R.-H. and J.L. Yordan, Zeta-potential measurements on microbubbles generated

using various surfactants, Journal of Colloid and Interface Science. 113(2), p. 430-438,

1986.

34. Li, C. and P. Somasundaran, Reversal of bubble charge in multivalent inorganic salt

solutions—Effect of magnesium, Journal of Colloid and Interface Science. 146(1), p. 215-

218, 1991.

35. Li, C. and P. Somasundaran, Reversal of bubble charge in multivalent inorganic salt

solutions—Effect of aluminum, Journal of Colloid and Interface Science. 148(2), p. 587-

591, 1992.

36. Graciaa, A., et al., The ζ-Potential of Gas Bubbles, Journal of Colloid and Interface

Science. 172(1), p. 131-136, 1995.

37. Creux, P., et al., Specific Cation Effects at the Hydroxide-Charged Air/Water Interface,

The Journal of Physical Chemistry C. 111(9), p. 3753-3755, 2007.

29

38. Yoon, R.H., D. Guzonas, and B.S. Aksoy, Role of Surface Forces in Tar Sand

Processing, in Proceeding of the 1st UBC-McGill Bi-Annual International Symposium

Vancouver, BC, Canada, 1995.

39. Yoon, R.H. and Y.I. Rabinovich, Role of Asphaltene in the Processing of Tar Sand, in

Proceeding of the 3rd UBC-McGill Bi-Annual International Symposium, CIM, Montreal,

Quebec, Canada, 1999.

40. Abraham, T., et al., Asphaltene-silica interactions in aqueous solutions: direct force

measurements combined with electrokinetic studies, Industrial & engineering chemistry

research. 41(9), p. 2170-2177, 2002.

41. Liu, J., et al., Colloidal interactions between asphaltene surfaces in aqueous solutions,

Langmuir. 22(4), p. 1485-1492, 2006.

42. De Gennes, P.-G., Scaling concepts in polymer physics, Cornell university press, 1979.

43. Patel, S.S. and M. Tirrell, Measurement of forces between surfaces in polymer fluids,

Annual Review of Physical Chemistry. 40(1), p. 597-635, 1989.

44. De Gennes, P., Polymers at an interface; a simplified view, Advances in Colloid and

Interface Science. 27(3), p. 189-209, 1987.

45. Berg, J.C., An introduction to interfaces & colloids: the bridge to nanoscience, p. 583,

World Scientific, 2010.

46. Alexander, S., Adsorption of chain molecules with a polar head a scaling description,

Journal De Physique. 38(8), p. 983-987, 1977.

47. Israelachvili, J. and R. Pashley, The hydrophobic interaction is long range, decaying

exponentially with distance, Nature. 300(5890), p. 341-342, 1982.

30

48. Pashley, R.M. and J.N. Israelachvili, A comparison of surface forces and interfacial

properties of mica in purified surfactant solutions, Colloids and Surfaces. 2(2), p. 169-

187, 1981.

49. Yoon, R.-H., D.H. Flinn, and Y.I. Rabinovich, Hydrophobic interactions between

dissimilar surfaces, Journal of colloid and interface science. 185(2), p. 363-370, 1997.

50. Parker, J.L., et al., Surface Forces between Plasma Polymer Films, Langmuir. 10(8), p.

2766-2773, 1994.

51. Yoon, R.-H. and S.A. Ravishankar, Long-Range Hydrophobic Forces between Mica

Surfaces in Alkaline Dodecylammonium Chloride Solutions, Journal of Colloid and

Interface Science. 179(2), p. 403-411, 1996.

52. Ishida, N., et al., Effects of Hydrophobizing Methods of Surfaces on the Interaction in

Aqueous Solutions, Journal of Colloid and Interface Science. 216(2), p. 387-393, 1999.

53. Binnig, G., C.F. Quate, and C. Gerber, Atomic force microscope, Physical review letters.

56(9), p. 930, 1986.

54. Ducker, W.A., Z. Xu, and J.N. Israelachvili, Measurements of Hydrophobic and DLVO

Forces in Bubble-Surface Interactions in Aqueous Solutions, Langmuir. 10(9), p. 3279-

3289, 1994.

55. Butt, H.-J., A Technique for Measuring the Force between a Colloidal Particle in Water

and a Bubble, Journal of Colloid and Interface Science. 166(1), p. 109-117, 1994.

56. Fielden, M.L., R.A. Hayes, and J. Ralston, Surface and Capillary Forces Affecting Air

Bubble−Particle Interactions in Aqueous Electrolyte, Langmuir. 12(15), p. 3721-3727,

1996.

31

57. Gillies, G., M. Kappl, and H.-J. Butt, Direct measurements of particle–bubble

interactions, Advances in Colloid and Interface Science. 114–115(0), p. 165-172, 2005.

58. Johnson, D.J., N.J. Miles, and N. Hilal, Quantification of particle–bubble interactions

using atomic force microscopy: A review, Advances in Colloid and Interface Science.

127(2), p. 67-81, 2006.

59. Ishida, N., Direct measurement of hydrophobic particle–bubble interactions in aqueous

solutions by atomic force microscopy: Effect of particle hydrophobicity, Colloids and

Surfaces A: Physicochemical and Engineering Aspects. 300(3), p. 293-299, 2007.

60. Nguyen, A.V., J. Nalaskowski, and J.D. Miller, A study of bubble–particle interaction

using atomic force microscopy, Minerals Engineering. 16(11), p. 1173-1181, 2003.

61. Englert, A., et al., Interaction forces between a deformable air bubble and a spherical

particle of tuneable hydrophobicity and surface charge in aqueous solutions, Journal of

colloid and interface science. 379(1), p. 121-129, 2012.

62. Ren, S., J. Masliyah, and Z. Xu, Studying bitumen–bubble interactions using atomic force

microscopy, Colloids and Surfaces A: Physicochemical and Engineering Aspects. 444 p.

165-172, 2014.

63. Derjaguin, B.V., Friction and adhesion IV. The theory of adhesion of small particles.,

Kolloid Z. 69 p. 155-164, 1934.

64. Israelachvili, J.N. and R.M. Pashley, Measurement of the hydrophobic interaction

between two hydrophobic surfaces in aqueous electrolyte solutions, Journal of Colloid

and Interface Science. 98(2), p. 500-514, 1984.

32

65. Claesson, P.M., et al., Interactions between water—stable hydrophobic Langmuir—

Blodgett monolayers on mica, Journal of Colloid and Interface Science. 114(1), p. 234-

242, 1986.

66. Eriksson, J.C., S. Ljunggren, and P.M. Claesson, A phenomenological theory of long-

range hydrophobic attraction forces based on a square-gradient variational approach,

Journal of the Chemical Society, Faraday Transactions 2: Molecular and Chemical

Physics. 85(3), p. 163-176, 1989.

67. Israelachvili, J.N., Intermolecular and surface forces: revised third edition, p. 158,

Academic press, 2011.

68. Nguyen, A. and H.J. Schulze, Colloidal science of flotation, p. 360, CRC Press, 2003.

69. Drost-Hansen, W., Ind. Eng. Chem. 61 p. 10, 1978.

70. Rabinovich, Y.I. and B. Derjaguin, Interaction of hydrophobized filaments in aqueous

electrolyte solutions, Colloids and Surfaces. 30(3), p. 243-251, 1988.

71. Li, Z. and R.H. Yoon, Thermodynamics of hydrophobic interaction between silica

surfaces coated with octadecyltrichlorosilane, Journal of colloid and interface science.

392 p. 369-375, 2013.

72. Wang, J., R.H. Yoon, and J.C. Eriksson, Excess thermodynamic properties of thin water

films confined between hydrophobized gold surfaces, Journal of colloid and interface

science. 364(1), p. 257-263, 2011.

73. Li, Z. and R.-H. Yoon, Thermodynamics of Solvophobic Interaction between

Hydrophobic Surfaces in Ethanol, Langmuir. 2014.

33

74. Parker, J.L., P.M. Claesson, and P. Attard, Bubbles, cavities, and the long-ranged

attraction between hydrophobic surfaces, The Journal of Physical Chemistry. 98(34), p.

8468-8480, 1994.

75. Stevens, H., et al., Effects of Degassing on the Long-Range Attractive Force between

Hydrophobic Surfaces in Water, Langmuir. 21(14), p. 6399-6405, 2005.

76. Tyrrell, J.W. and P. Attard, Images of nanobubbles on hydrophobic surfaces and their

interactions, Physical Review Letters. 87(17), p. 176104, 2001.

77. Wang, J. and R.-H. Yoon, AFM Forces Measured between Gold Surfaces Coated with

Self-Assembled Monolayers of 1-Hexadecanethiol, Langmuir. 24(15), p. 7889-7896,

2008.

78. Zhang, J.H., et al., Effects of degassing and ionic strength on AFM force measurements in

octadecyltrimethylammonium chloride solutions, Langmuir. 21(13), p. 5831-5841, 2005.

79. Meyer, E.E., Q. Lin, and J.N. Israelachvili, Effects of Dissolved Gas on the Hydrophobic

Attraction between Surfactant-Coated Surfaces, Langmuir. 21(1), p. 256-259, 2004.

80. Tsao, Y.H., et al., Interactions between hydrophobic surfaces. Dependence on

temperature and alkyl chain length, Langmuir. 7(12), p. 3154-3159, 1991.

81. Kekicheff, P. and O. Spalla, Refractive Index of Thin Aqueous Films Confined between

Two Hydrophobic Surfaces, Langmuir. 10(5), p. 1584-1591, 1994.

82. Tsao, Y.H., D.F. Evans, and H. Wennerstroem, Long-range attraction between a

hydrophobic surface and a polar surface is stronger than that between two hydrophobic

surfaces, Langmuir. 9(3), p. 779-785, 1993.

83. Christenson, H.K. and P.M. Claesson, Cavitation and the interaction between

macroscopic hydrophobic surfaces, Science. 239(4838), p. 390-392, 1988.

34

84. Ruckenstein, E. and N. Churaev, A possible hydrodynamic origin of the forces of

hydrophobic attraction, Journal of Colloid and Interface Science. 147(2), p. 535-538,

1991.

35

Chapter 2

Temperature Effect on the Stability of Wetting Films of

Water on Bitumen

Abstract

For air bubbles to attach on mineral (or bitumen) particles, the wetting films of water

formed on the surfaces must rupture. Thermodynamically, the films can rupture when the

disjoining pressures () of the films are negative. In the present work, the disjoining pressures of

the wetting films formed on bitumen have been measured at temperatures in the range of 22 to

80 °C. The results show that becomes less negative with decreasing temperature. However, it

remains negative until the temperature is decreased to 35 °C, but becomes positive at 22 °C. These

results are consistent with the industrial experiences in bitumen flotation.

2.1 Introduction

In 2012, Canada produced 1.8 million bbl/day of oil from the oil sands resources in Alberta,

which is expected to grow to 5.2 million bbl/day by 2030, according to the projections from the

Canadian Association of Petroleum Producers (CAPP). The amount of heavy oil (bitumen)

deposited in the form of oil sands is estimated to be 2.5 trillion bbls, which is five-times larger

than the conventional oil reserves in Saudi Arabia. At present, approximately 55% of the bitumen

36

is produced from mining followed by flotation, with the rest produced by in-situ extraction

methods, e.g., cycle steam stimulation (CSS) and steam-assisted gravity drainage (SAGD).

The oil sands ores are composed of sand grains with a median size of 65 mesh (208 µm)

and a top size of 14 mesh (1,168 µm), a layer of hydrated water (or innate water), and bitumen that

forms a continuous phase between sand grains.3 The innate water contains ultrafine clay particles

such as kaolinite and illite. The ores containing typically 7 to 14% by weight of bitumen are

processed by flotation, with recoveries usually in the range of 93%, while those of the in situ

processes are much lower (40 to 70%). In general, the ores containing large amounts of ultrafine

particles are of low grades and suffer from low recoveries. Also, the ores containing divalent

cations such as Ca2+ and Mg2+ ions are difficult to process.5

The basic bitumen extraction processes used today are the same as the original hot-water

processing method developed by Clark in the 1920s.7 In this process, bitumen is detached (or

liberated) from sand grains by agitating an aqueous ore slurry in hot water (~80 °C) in the presence

of caustic soda to raise the pH to 8.0 – 8.5. Initially, the agitation was provided by a tumbler along

with steam injection. In early 1990’s, the tumblers have been replaced by pipelines, in which

bitumen is liberated and aerated in transit. The aerated bitumen droplets float in a primary

separation vessel (PSV) to be separated from the sand grains and clays. The middlings that are

insufficiently aerated are directed to a bank of conventional and/or column flotation cells to recover

additional bitumen. The froth products, containing typically 60% bitumen, 30% water and 10%

solids, are de-aerated, diluted in naphtha (or n-alkane), and centrifuged to remove solids. The clean

bitumen is then upgraded by a combination of thermal and catalytic hydrocracking processes to

obtain synthetic crude oils (SCO).

37

Much work has been done to lower operating temperatures to reduce the energy

consumption. In 1990, Sury9 developed a cold water process for bitumen extraction, in which an

oil sand matrix was attrition-scrubbed in water in the presence of appropriate conditioning agents

to achieve liberation and disperse the liberated bitumen. The aqueous slurry was then subjected to

flotation. The conditioning agents included typical flotation reagents such as kerosene, diesel oil,

and frother, e.g., methyl-isobutyl-carbinol (MIBC). Bitumen recoveries were ~90 % at 5 to 10 °C

and ~95% at 15 to 35 °C. In 2000, a similar process was tested at the Aurora plant, Syncrude

Canada, Ltd., at 25 °C.10 In 2002, the operating temperature was raised to 35 to 40 °C to obtain

higher and more consistent recoveries.

Despite the simplicity in the extraction process and the significant improvement achieved

in recent years, some of the mechanisms involved in the different steps of the extraction process

are not fully understood. Contrary to the general belief that bitumen-in-water emulsions are

stabilized by the surfactants that are naturally present in bitumen (e.g., naphthenate), it has been

shown that the emulsions are stabilized by the asphaltene polymers present in bitumen.11 This

mechanism was proposed on the basis of the surface forces measured between two bitumen-coated

mica surfaces using a surface force apparatus (SFA). Recognizing the role of asphaltene as an

emulsifier may help explain the long conditioning times and high temperatures required for

bitumen extraction.

In general, the bubble-bitumen attachment (or aeration) requires long induction times

usually in the range of seconds, while the bubble-particle attachment in mineral flotation occurs in

milliseconds.11-13 Further, air bubbles attach on mineral surfaces, forming finite contact angles,

while in bitumen flotation small bubbles are engulfed into bitumen droplets. Despite the

differences noted above, the fundamental mechanisms are the same in that the wetting films of

38

water formed on the substrates (mineral and bitumen) must be broken for the flotation to occur. It

is more difficult for air bubbles to break the wetting films formed on bitumen than those formed

on mineral most probably due to the presence of asphaltene polymers exposed on the former.

When an air bubble is pressed against the surface of a bitumen droplet (or a mineral particle)

under turbulent conditions, it deforms; and the curvature change associated with the deformation

creates an excess pressure (p), which is equal to 2/R, where is the surface tension and R is the

radius of curvature. The excess (or Laplace) pressure causes the thin liquid film (TLF) of water

(or wetting film) formed between the two macroscopic surfaces to thin. As the film thins further,

the surface forces in the film come into play and create a disjoining pressure () at h < 200 to 300

nm, in which case the excess pressure should become,

R

p2

[2.1]

Eq. [2.1] shows that the excess pressure becomes large when < 0, which will accelerate the film

thinning and cause the film to rupture at a critical thickness (hcr). This will result in bubble-bitumen

attachment and subsequent engulfment. If > 0, however, the wetting film is stable and the

bubble-particle attachment becomes difficult.

The highest value of in a disjoining pressure isotherm ( vs. h curve) may represent the

energy barrier (E1) for bubble-particle attachment. In bitumen flotation, E1 is high most probably

due to the repulsive steric force created by the asphaltene exposed on bitumen surface11 and due

to the high negative -potentials created by the alkali (KOH) added to the system to help liberate

the bitumen from sand grains. According the bubble-particle attachment model proposed by

Luttrell and Yoon (1992),9

k

aE

EP 1exp [2.2]

39

it is necessary to apply a high kinetic energy (Ek) to overcome the energy barrier and achieve a

high probability of attachment (Pa). An alternative to applying a high Ek would be a long

conditioning time, which may be provided by the pipeline transport (or hydrotransport) system

used in industry today.

Still another alternative is to reduce the energy barrier (E1) by appropriate means. It has

been shown that the induction time for bubble-bitumen interaction decreases at lower pH,11

suggesting a corresponding decrease in E1. However, flotation at low pH is impractical in a sense

that bitumen liberation requires a high pH. In the present work, the effects of temperature on E1

have been studied by measuring at temperatures in the range of 22 to 80 °C. The measurements

were conducted using the modified thin film pressure balance (TFPB) technique described

previously.10,11 Since this technique is capable of measuring pressures in unstable films, the

measurements have been extended to the TLFs of water formed on hydrophobic substrates such as

maltene and bitumen at elevated temperatures of up to 80 °C. The results are in good agreement

with industry practice in terms of temperature effect on bitumen extraction. It is hoped that the

results of the present work will be useful for better understanding the role of temperature in

bitumen extraction by flotation.

2.2 Experimental

2.2.1 Materials

A sample of bitumen (99 wt.%) was provided by Suncor Energy, Inc. Reagent grade

toluene (>99.5% purity) was purchased from Spectrum Chemical Mfg. Corp., and HPLC grade n-

heptane (99.3% purity) was obtained from Fisher Scientific. Millipore ultrapure water with a

resistivity of 18.2 MΩ·cm was obtained using a Direct Q3 water purification system. Silicon

wafers (100 crystal planes, University Wafer) were used as the substrates for coating bitumen,

40

asphaltene and maltene. Sulfuric acid (H2SO4, 98% purity, VMR international) and hydrogen

peroxide (H2O2, 29.0–32.0% purity, Alfa Aesar) were used to clean the silicon substrates.

2.2.2 Separation of Asphaltene and Maltene

The bitumen sample was dissolved in toluene at a toluene/bitumen ratio of 5:1 by volume.

The solution was then stirred using a magnetic stirrer for 2 hr, centrifuged for 30 min to remove

the solids, and subsequently put under a fume hood to allow the toluene to naturally evaporate.

After the toluene evaporation was complete, n-heptane was added to the toluene- and solids-free

bitumen at a heptane/bitumen ratio of 40:1 by volume. The n-heptane and bitumen mixture were

stirred by means of a magnetic stirrer for 2 hr, during which time maltene dissolved into n-heptane

while asphaltene precipitated out. The mixture was left overnight for the asphaltene to settle at the

bottom. The supernatant n-heptane was carefully removed by means of a pipette and placed under

a fume hood to let the n-heptane naturally evaporate, leaving maltene behind. The amount of the

evaporated n-heptane was small.

The asphaltene precipitate was repeatedly washed for 17 times with an excess amount (~1L)

of n-heptane to remove any maltene that may have co-precipitated. In each washing step, the

mixture was agitated for 2 hr and left to stand overnight to allow the asphaltene to precipitate and

the supernatant was carefully removed. After the last washing step, the supernatant n-heptane

became colorless. The asphaltene precipitate was left under a fume hood to allow n-heptane to

evaporate and thereby obtain a sample of purified asphaltene.

2.2.3 Preparation of Bitumen-, Maltene-, and Asphaltene-coated Surfaces

Silicon wafer was first oxidized in an oxidation furnace at 1070 °C for ~40 min using wet

oxidation procedure. The oxidized silicon wafers were then cut into 12x12 mm pieces and used as

substrates for bitumen, asphaltene, and maltene. Before the coating, the silicon substrates were put

41

in an ultrasonic bath of water for ~1 hr and then cleaned by immersing in a Piranha solution (3:7

by volume of H2O2/H2SO4) at 120 °C for 1 hr, followed by rinsing with an adequate amount of

ultrapure water and pure nitrogen (N2) blow-drying. The silicon substrate was placed in a spin-

coater (Laurell, WS-400-6NPP), rinsed with toluene, and then spun at 6,000 rpm for 30 s to dry

off the toluene. When the substrate was free of toluene, a small amount of (2 or 3 drops) of a

bitumen-in-toluene solution (2.5 mg/mL) was placed on the substrate and let it stand for 10 s. The

substrate was then spun at 3,500 rpm for 20 s before placing another drop of the bitumen solution

and continuing spinning for additional 10 s. The spinning continued for an additional 1 minute at

6,000 rpm to evaporate the solvent and obtain a smooth and uniform bitumen coating. The spin-

coated bitumen was then placed in a particulate matter-free laminar hood for ~30 min to ensure

maximum removal of toluene from the coating. The procedure described above was also employed

for the preparation of asphaltene- and maltene-coated silicon surfaces.

2.2.4 Measurement of Film Thinning Kinetics

Figure 2.1 shows the modified thin film pressure balance (TFPB) apparatus used in the

present work to study the kinetics of film thinning. The original TFPB apparatus was designed for

the study of foam films (or the water films between two air bubbles) by Scheludko and Exorowa.14

The equipment was modified to study the kinetics of film thinning for wetting films.15 As shown

in the inset, a flat surface of interest is placed on the top of a film holder (2.0 mm diameter glass

tubing) with its surface facing downward, so that the surface is in contact with the water in the

film holder. In the present work, the flat silicon wafers coated with bitumen, asphaltene, or maltene

were used as samples of interest.

In a given experiment, the silicon wafer-glass tubing assembly was inserted in a glass cell

surrounded by a water jacket, in which water of known temperature was circulated. The film

42

holder-water jacket assembly was placed on an inverted microscopic stage (Olympus IX51) to

monitor the changes in film thicknesses as a function of time. A mercury arc short lamp (100 W,

Olympus) was used as a light source with a bandpass interference filter (#65-643, Edmund Optics)

to produce a monochromatic green light source (λ=546 nm). In a given experiment, the liquid in

the film holder is slowly withdrawn by turning the knob of the piston. Once the interference

patterns (Newton rings) began to appear in the microscopic field of view, the film was allowed to

thin spontaneously while recording the images by means of a high-speed camera (Fastec Imaging,

HS400115). The interference patterns recorded were used to reconstruct timed thickness profiles

of a wetting film across the entire film holder with a nanometer resolution. The spatial and

temporal film profiles of the wetting film were then used to monitor the changes in film thicknesses

(h) as a function of time and radial distance of the film at any point of the film. The experimental

Figure 2.1 Schematics of the modified TFBC apparatus used for the study of wetting

films.

43

data were used to obtain information on the film thinning kinetics and calculate the disjoining

pressure () in the thin liquid films (TLFs) according to Eq. [2.3], as derived by Pan. et.al.,16

r

r

r

rdrdr

t

hr

rhr

hr

rrR 03

112

2

[2.3]

where is the air/water interfacial tension, R the bubble radius, r the radial position, h the wetting

film thickness, and µ the fluid viscosity. The experiments were conducted at temperatures in the

range of 22 to 80 °C.

2.2.5 Contact Angle Measurements on Bitumen Film

Flat plates of silicon wafers were coated with bitumen using the spin-coating technique

described in the previous section (2.2.3). The thicknesses of the coatings were approximately 11

nm. The hydrophobicity of the bitumen-coated surfaces were determined by measuring water

contact angles using the captive bubble technique. For each measurement, an air bubble was

brought to a bitumen-coated surface from underneath. The measurements were conducted using a

Ramé-Hart contact angle goniometer, Model 250, F4 series. The measurements were conducted in

a double-jacketed rectangular glass cell with optical windows filled with ultrapure water. The

measurements were conducted at different temperatures in the range of 22 to 80 oC under

controlled temperature conditions.

Effect of temperature was also studied by measuring contact angles on relatively thick

(0.5 mm) films of bitumen formed on clean Teflon plates. Several bitumen-coated Teflon plates

were placed in an oven for 3 to 4 min at 120 °C to soften the bitumen and obtain flat surfaces,

which were subsequently cooled at room temperature in a dust-free cabinet. To measure contact

angles at an elevated temperature, a bitumen-coated Teflon plate was placed in an oven at a desired

temperature for a few minutes and then placed in ultrapure water at approximately the same

44

temperature. The water temperature was controlled by circulating the water from a constant

temperature bath through a double-walled glass cell that was designed for captive bubble contact

angle measurement. The Ramé-Hart contact angle goniometer was used to record the images of

air bubbles interacting with bitumen-coated surfaces at a speed of 30 fps and to determine dynamic

contact angles offline using a protractor.

2.3 Results and Discussion

2.3.1 Temperature Effect on the Film Thinning Kinetics

Figure 2.2 shows the changes in Newton rings (or interference patterns) of the wetting films

formed on bitumen-, asphaltene-, and maltene-coated silicon surfaces obtained at 80 °C. As shown,

the Newton rings changed with time due to film thinning until the film ruptured. From the light

intensities of the interference patterns, one can calculate the film thickness (h) using the micro-

interferometric technique and obtain temporal and spatial profiles of the wetting films, as shown

Figure 2.2 Changes of Newton rings with time due to the decrease in the thickness of the

wetting film of water formed on maltene (a), bitumen (b) and asphaltene (c) surface, while

the air bubble was approaching the substrate at 80 °C.

45

in Figure 2.3. The arrow in each profile represents the film thickness at which the wetting film

ruptured catastrophically.

0

200

400

600

(c)

(b)

0.12

0.08

0.04

0 s

80 oC

h (

nm

)

Maltene

0

200

400

600

0.16

0.1

0.05

0 s

h (

nm

)

80 oC

Bitumen

-40 -20 0 20 400

200

400

600

0.61

0.3

0.1

0 s

h (

nm

)

r (m)

80 oC

Asphaltene

-40 -20 0 20 40

(a)

Figure 2.3 Spatial and temporal profiles of the wetting films of water formed on maltene,

bitumen, and asphaltene surface, respectively, at 80 °C. The arrows represent the critical

thicknesses where the films rupture.

46

As shown in Figure 2.3, the films formed on maltene and bitumen surfaces thinned fast and

ruptured at 0.12 and 0.16 s, respectively. On asphaltene, the film thinned substantially slower, with

the rupture occurring at 0.615 s. Note here that the wetting film before rupture formed on

asphaltene were flatter than those formed on the other two substrates most probably due to the

presence of a greater resistance force to film thinning.

Figure 2.4 shows the effect of temperature on the kinetics of film thinning on bitumen. The

film thicknesses (h) measured at the film centers, i.e., r = 0, are plotted vs. time (t) at different

temperatures. As shown, the film thinning kinetics represented by the slope of the h vs. t curves

increased with increasing temperature, which explains the advantage of floating bitumen at higher

temperatures. According to the kinetics curves presented in Figure 2.4, bitumen flotation would

be possible at temperatures as low as 35 °C. At 22 °C, it took >80 s for the wetting film to rupture,

0 4 8 120

200

400

Natural pH

55

22 0C

8068

35

h (

nm

)

t (s)

Bitumen

22 0C

80 6855

35

0 40 800

100

200

300

h (

nm

)

t (s)

Figure 2.4 Effect of temperature on the kinetics of film thinning on bitumen in water at natural

pH of about 5.7.

47

which would make it difficult to float bitumen at such a low temperature. The results presented in

Figure 2.4 are consistent with the industrial experiences reported in the literature. The original hot

water extraction process was designed to work at 80 to 85 °C. During the 1990s, the operating

temperatures were reduced to 70 to 80 °C to save energy. In 2000, the Aurora plant of Syncrude

Canada, Ltd., tested the cold water processing at approximately 25 °C.10 However, the operating

temperature has been increased to 35 to 40 °C, which is consistent with the results obtained in the

present work. At present, most plants operate at temperatures in the range of 40 to 55 °C to ensure

operation reliability and high bitumen recovery. The results that the film thinning became faster at

higher temperatures are also in agreement with the induction time measurements reported by Gu

et al.,12, 17 who found that the induction time of bubble-bitumen attachment decreased greatly with

increasing temperature.

Figures 2.5 and 2.6 show the h vs. t curves for the wetting films formed on maltene and

asphaltene, respectively, at temperatures in the range of 35 to 80 °C. As shown, the kinetics of film

thinning increased with increasing temperature. Also, the thin liquid film of water thins much

faster on maltene than on asphaltene at a given temperature, indicating a greater resistance on

asphaltene surface.

Comparison of the results presented in Figures 2.4, 2.5, and 2.6 shows that the kinetics of

film thinning increased in the order of asphaltene, bitumen, and maltene, which in turn suggests

that the hydrophobicity of the three different materials increases in the same order. This finding is

in good agreement with the work by Yoon and Rabinovich,18 who conducted direct force

measurements using an atomic force microscope (AFM) between bitumen, asphaltene and maltene

surfaces in water and found that there was a long-range attractive force between maltene surfaces

48

0 4 8 120

100

200

300h (

nm

)

80

68

55

t (s)

35 0C

Natural pH

Maltene

Figure 2.5 Effect of temperature on the kinetics of film thinning on maltene.

0 4 8 120

200

400

h (

nm

)

80

68

55 35 0C

35 0C

Natural pH

80

68

55

t (s)

Asphaltene

0 40 800

150

300

h (

nm

)

t (s)

Figure 2.6 Effect of temperature on the kinetics of film thinning on asphaltene

49

while a strong repulsive steric force between asphaltene surfaces. The steric force may be

responsible for the slow kinetics of film thinning on asphaltene-coated surfaces.

2.3.2 Temperature Effet on Disjoning Pressure

Figure 2.7 shows the disjoining pressure () in the wetting films of water formed on

maltene, asphaltene, and bitumen surface at temperatures in the range of 20 to 80 °C. The

disjoining pressure was calculated using a spatial and temporal profile obtained in the present work.

Figure 2.3 shows the data obtained at 80 °C, while the profiles obtained at other temperatures are

not shown in this communication. The film profiles were used to derive the kinetic information on

film thinning, i.e., 𝜕ℎ 𝜕𝑡⁄ and 𝜕ℎ 𝜕𝑟⁄ , which were then substituted into Eq. [2.3] and obtain . As

shown in Figure 2.7, both maltene and bitumen exhibited negative disjoining pressures ( < 0) at

temperatures between 35 to 80 °C. On asphaltene, > 0 at all temperatures except at 80 °C. These

results suggest that maltene and bitumen can be more readily floated than asphaltene. In general,

decreases with increasing temperature, which is consistent with the industrial experience that

bitumen flotation becomes easier at higher temperatures.

2.3.3 Contact Angle Measurements at Different Temperatures

The results obtained in the present work suggest that the hydrophobicity of bitumen

increased with increasing temperature. In order to better understand this observation, the contact

angles of water on bitumen-coated surfaces, on both thin and thick bitumen films, were measured

at different temperatures using the captive bubble method. Figure 2.8 shows the dynamic contact

angles measured on thin bitumen films (11 nm thick) formed on silicon wafer at different

temperatures. At a given temperature, the contact angle decreased with time, as was also observed

by others.19 Note here that contact angles decreased with increasing temperature, which is contrary

50

-1500

-1000

-500

0

80

68 55 35 OC

Natural pH

Asphaltene

-1500

-1000

-500

0

(

Pa)

(

Pa

)

35 OC

55

80

68

(P

a)

Natural pH

Maltene

0 100 200 300

-1500

-1000

-500

0

(c)

(b)

80

68

5535

20 OC

h (nm)

Natural pH

Bitumen

0 100 200 300

(a)

Figure 2.7 Effect of temperature on the disjoining pressure isotherms ((h)) in the thin films

of water formed on maltene (a), asphaltene (b), and bitumen (c). The data have been collected

at the center of films.

51

to what is expected in view of increasing flotation recovery with temperature.20 It is also contrary

to our findings that film thinning kinetics on bitumen increased with temperature and the disjoining

pressure became more negative at a higher temperature.

The decrease in contact angle with time may be attributed to changes in the orientation of

the surfactants present in bitumen, as suggested by Ren, et al.19 It is more likely, however, that the

bitumen films used for the contact angle measurement were so thin that parts of the films dissolved

away into water and, thereby, exposed the hydrophilic silicon dioxide substrates. The higher the

temperature, the faster and more bitumen would dissolve into the solution. Therefore, a series of

contact angle measurements were conducted on a thick bitumen film of 0.5 mm to avoid the

exposure of silicon dioxide substrates toward water. Figure 2.9 shows the images of the air bubbles

adhering on bitumen surfaces at different temperatures, while Figure 2.10 shows the dynamic

contact angles measured from the images by extrapolating the curvatures of the bubbles. A close

0 20 40 600

30

60

90

(

o )

t (min)

68

55 35

22 oC

Figure 2.8 Dynamic water contact angles measured on the thin bitumen films at

different temperatures. The thickness of the bitumen films were 11 nm.

52

Figure 2.9 The kinetics of bubble-bitumen interactions at 22, 35, and 80 oC. The

thickness of the film was about 0.5 mm.

0 10 20 30 40

60

90

120

150

22

35

80 °C

°

Time (s)

Figure 2.10 Dynamic contact angles on the thick bitumen films at different temperatures in

water. The measurement was difficult due to the liquid (oil) collecting around the three-

phase contact lines as shown in Figure 2.9. Therefore, the contact angles were measured by

extending the bubble surface along the curvatures.

53

examination of the images shows that a type of liquid, most probably maltene, collects near the

three-phase contact points, and the amount of the liquid increases with temperature.

When bitumen is exposed to water, asphaltene (or the surfactant naturally present in

bitumen) may be exposed to the surface, rendering the surface less hydrophobic. As the

temperature increases, maltene will diffuse through the asphaltene layer and restore the

hydrophobicity of the bitumen surface, causing the contact angle to increase, as shown in Figure

2.10. Furthermore, part of the air bubble may be covered by maltene. As shown, the contact angles

increased with time and temperature, indicating that the diffusion of maltene is slow at room

temperature, but the rate of diffusion increases with temperature due to a corresponding decrease

in viscosity.

Figure 2.11 depicts a model showing how bitumen contact angles increase with time at a

high temperature. Initially, the contact angle is relatively low due to the presence of asphaltene on

the surface, which is hydrophilic at an alkaline pH in the range of 8 to 10. At a pH close to its

pKa, its functional groups are ionized and strongly interact with water (or become hydrophilic). At

an elevated temperature, maltenes diffuse into the asphaltene layer on the surface and increase its

Figure 2.11 A model for the attachment of air bubble to the surface of bitumen at a high

temperature: a) initially, an air bubble attaches to a bitumen surface coated by asphaltene

with a low contact angle; b) maltene diffuses into the asphaltene layer on the surface and

increases the hydrophobicity and hence the contact angle. This is possible due to the low

viscosity of maltene at higher temperatures. Eventually, the bitumen layer saturated with

maltene begins to cover the bubble surfaces.

54

hydrophobicity, causing the contact angle to increase, as shown. (Recall that in the present work

the contact angle measurements began as soon as a sample was placed in water.) When the

asphaltene layer is saturated with maltene, an oily substance begins to coat the bubble surface, as

shown in the images of the dynamic contact angle measurements (see Figure 2.9). At the same

time, the contact angles increase, and the profiles of the bubbles change accordingly. As the bubble

profiles become flat, the bubble-bitumen contact area increases, which may facilitate the diffusion

of air into bitumen underneath through the asphaltene layer. The time it takes for the contact angle

to increase and reach an equilibrium decreases with increasing temperature most probably due to

the decrease in viscosity. It can be seen from Table 2.1 that the dynamic viscosity of bitumen and

maltene decreased dramatically with temperature. For example, the viscosity of Athabasca

bitumen decreases by nearly three orders of magnitude as temperature increases from 20 to 75 °C.

The viscosity data presented in Table 2.1 were experimentally measured using different types of

Table 2.1 Dynamic Viscosity Data for Bitumen, Maltene, n-Dodecane, Hexacane and

Naphthalene.

Viscosity (mPa·s)

20 °C 30 °C 40 °C 50 °C 60 °C 75 °C

Athabasca bitumen1 7x105 ~105 ~4x104 ~104 ~5x103 ~103

Lloydmin bitumen2 38601 11591 3999 1604 746

Maltene extracted from

Lloydmin bitumen2 2585 1017 431 219 119

n-Dodecane (C12H26)4 ~1.344 0.911 0.659

Hexadecane (C16H34)6 ~3.061 1.829 1.231

Naphthalene (C10H8)8 2.483 1.422

55

viscometers by previous researchers. Details about the viscometers used could be found in the

cited papers.1, 2, 4, 6, 8 The increase in contact angle with time and temperature, as shown in Figure

2.10, may be attributed qualitatively to the increased saturation of the asphaltene layer with

maltene at longer contact times and higher temperatures. Furthermore, the hydrophobicity of

maltene itself increases with temperature, as shown in the disjoining pressure measurements (see

Figure 2.7a).

2.3.4 Thermodynamic Analysis Based on Acid-base Theory

As noted in the foregoing paragraph, it appears that the oily substance coats the bubble

when the asphaltene layer is saturated with maltene. It was thought first that maltene could readily

coat bubble surface due to the hydrophobic interaction between maltene and air bubble as both are

hydrophobic. A simple thermodynamic analysis suggests, however, that maltene is less likely to

coat the bubble than the asphaltene saturated with maltene (or bitumen). The free energy change

associated with the spreading of an oily substance on an air bubble ( sG ) can be obtained using

the following relation,

wwoosG / [2.4]

where o is the surface tension of oil, wo / is the interfacial tension between oil and water, and

w is the surface tension of water. Eq. [2.4] shows that the spreading of an oily substance on an

air bubble entails the expansion of the oil/air and oil/water interfaces at the expense of the air/water

interface of the bubble. In the present work, maltene may be considered the oily substance.

According to the acid-base theory,21

wowo

d

w

d

owowo 222/ [2.5]

56

where d

o and d

w are the dispersion components of the oil and water interfaces, respectively;

o

and

w are the acidic components of oil and water, respectively;

o and

w are the basic

components of oil and water, respectively.

Substituting Eq. [2.5] into Eq. [2.4], one obtains,

wowo

d

w

d

oosG 2222 [2.6]

If maltene is as hydrophobic as other hydrocarbon oils , o and d

o should be in the range of 17 to

25 mJ/m2, while d

w = 21.8 mJ/m2,

w =

w = 25.5 mJ/m2, and

o and

o are approximately zero,

assuming maltene is completely apolar.22 Assuming also that o = 17 mJ/m2 for maltene, Eq. [2.6]

gives sG = -4.5 mJ/m2, suggesting that maltene can coat the bubble surface. If o = 25 mJ/m2

rather than 17 mJ/m2, the free energy change becomes positive, and thus maltene cannot coat air

bubbles. Since the surface tensions of oils decrease with increasing temperature in general, it is

thermodynamically possible that maltene can coat air bubbles at relatively high temperatures. It is

unfortunate that detailed thermodynamic information for maltene is not available in the literature.

The authors contacted the University of Alberta looking for a correct value of o for maltene to

no avail. The data do not exist as yet. It would be fruitful to determine the missing thermodynamic

data in the future.

The surface tension data for bitumen are better known than those for maltene. According

to the data reported by Moran et al.,23 o = 30 mJ/m2 for bitumen, and wo / are in the range of 11

to 20 mJ/m2. Substituting these values into Eq. [2.4] along with w = 72 mJ/m2, one obtains that

sG is in the range of -22 to -31 mJ/m2. It appears, therefore, that bitumen (or asphaltene saturated

with maltene) is more likely to coat air bubbles during bitumen flotation. The reason for this is that

57

the polar groups of asphaltene can interact with the surrounding water molecules and give rise to

a more negative free energy of spreading ( sG ), as is obvious from Eq. [2.6]. Conceptually, a

hydrophobic oil will not spread at the air/water interface or bubble surface because it cannot

interact with the water surrounding bubbles. If a surfactant is added to the oil, however, it can

spread because the polar heads of the surfactant molecules interact with the surrounding water

molecules and reduce free energy. Asphaltene acts effectively as a surfactant in the bitumen

extraction system and facilitates the spreading of bitumen (or maltene) on bubble surfaces. As

mentioned earlier, asphaltene contains polar groups such as carboxyl, sulfide, amide, and hydroxyl

groups. In the spreading process, asphaltene molecules in bitumen tend to adsorb at the air/water

interface because of these polar groups, lowering the bitumen/water interfacial tension and hence

the Gibbs free energy change for spreading. Asphaltene acts as a spreading agent for bitumen.

It should be noted here that maltene is not completely apolar because one of its components,

resin, contains polar groups such as carboxylates and sulfates.24 Compared to asphaltene, however,

resin is less polar.25 Asphaltene still plays a major role in facilitating the spreading of bitumen onto

air bubbles.

2.3.5 A Bitumen Aeration Model

Based on the contact angle measurements shown in Figures 2.9 and 2.10 and the

thermodynamic considerations discussed above, a conceptual model for bitumen aeration may be

suggested as depicted in Figure 2.12. When a small air bubble collides with a bitumen droplet in a

hydrotransport pipeline or in a secondary flotation cell, the bubble will deform and create an excess

pressure (p) in the wetting film of water formed in between the bitumen droplet and the air bubble.

At this point in time, the film will thin due to the curvature pressure (pcur). When the thickness

reaches the range of 250 to 300 nm, the film thinning process may encounter a resistance from the

58

colloidal forces in the form of a positive disjoining pressure ( > 0). The film thinning will stop

when pcur = , because the excess pressure in the film becomes zero. If < 0, on the other hand,

the excess pressure will increase, as p = pcur - , and hence accelerate the film thinning process

and eventually rupture the wetting film. Once the film has ruptured, a contact angle is formed.

Initially, the contact angle is small as depicted in Figure 2.12a. The experimental data

presented in Figures 2.9 and 2.10 show actually that initial contact angles are less than 90°. In

time, however, contact angles increase above 90o, as depicted in Figure 2.12b. According to the

Young’s equation,

w

wbb

/cos

[2.7]

the contact angle increase should be due to the changes in interfacial tensions involved, Eq. [2.7]

suggests that contact angle () should be less than 90o when wbb / . The contact angle should

increase above 90° when wbb / . Therefore, the dynamic contact angle behavior of bitumen as

manifested in Figures 2.9 and 2.10 must be due to the changes in interfacial tensions. It should be

Figure 2.12 A dynamic bitumen aeration model: a) an air bubble attaches to an asphaltene-

coated bitumen surface, forming a low contact angle; b) maltene diffuses into the asphaltene

layer on the surface and increases the contact angle of the layer, and subsequently covers bubble

surface; c) when the air bubble is covered by the maltene-saturated asphaltene, the bubble enters

bitumen droplet, completing aeration. Thermodynamically, maltene-saturated asphaltene (or

bitumen) spreads on air bubble better than maltene. Thus, asphaltene may act as a spreading

agent for maltene.

59

noted also that the secret of achieving high contact angles, which is essential for successful

aeration, is to increase the bitumen/water interfacial tension ( wb / ).

As has already been noted, it is necessary to increase the disjoining pressure for bitumen

liberation, which can be achieved by agitating a bitumen slurry at a reasonably high temperature

in the presence of an alkali (KOH). Under this condition, the basic functional groups of asphaltene,

which includes carboxylic acids, carbonyls, phenols, pyroles, and pyridinic functional groups,

dissociate and become polar.26-29 As a result, highly-charged asphaltene molecules are exposed on

bitumen surface and create steric and electrical double-layer forces to create a highly-positive

disjoining pressure in the wetting films of water between bitumen and silica sands. At the same

time, the exposure of the polar asphaltene should render the bitumen surface polar by increasing

the acidic and basic components of surface tensions, i.e.,

b and

b . This should in turn decrease

the bitumen/water interfacial tension ( wb / ) according to the acid-base theory,

wbwb

d

w

d

bwbwb 222/ [2.8]

After bitumen is liberated from sand grains, bitumen aeration is desired. The aeration of

bitumen consists of two substeps: 1) the attachment of bitumen droplets to air bubbles and 2) the

subsequent engulfment of air bubbles by bitumen. The bubble-bitumen attachment, like in the case

of mineral flotation, involves the thinning of the wetting film of water formed between bitumen

droplets and air bubbles and the subsequent film rupture at a critical thickness forming a three

phase contact line. The bubble engulfment involves the spreading of the bitumen onto air bubbles.

Attachment (or film rupture) is the prerequisite of the engulfment. Without the film rupture,

engulfment will not happen and the bitumen flotation is not possible. If a bitumen droplet and an

60

air bubble collide, their mutual attachment is thermodynamically favorable if the Gibbs free energy

change associated in the attachment (Ga) is negative. Ga can be given as the following equation,

)( / wwbbaG [2.9]

which represents a process of a new air/bitumen interface being created at the expense of the

air/water and bitumen/water interfaces. If Ga is negative, the attachment is favored and

spontaneous. If Ga is positive, the bitumen should not attach. From Eq. [2.9], one can see that

high bitumen/water interfacial tension ( wb / ) will favor the bitumen-air attachment

For bitumen-bubble attachment, it is necessary to render the disjoining pressure in the

wetting films formed between air bubble and bitumen negative, for which the asphaltene molecules

exposed on bitumen surface need to become less polar so that the bitumen/water interfacial tension

is increased. The most likely way to achieve this goal will be to allow the maltene at the interior

of a bitumen-in-water emulsion droplet to diffuse out to the exterior through the asphaltene layer

exposed on the surface and render the droplet surface hydrophobic. According to Eq. [2.8], one

must reduce the acidic and basic components of bitumen surface tension, which can be readily

accomplished when the maltene exposed on the surface masks the polar groups of the asphaltene

on the surface. Once this has been accomplished, air bubble begins to penetrate the asphaltene

droplet, as shown in Figure 2.12b, with the asphaltene acting effectively as an emulsifier for

maltene, as has already been discussed.

In effect, bitumen droplets rendered hydrophilic during the liberation step are rendered

hydrophobic in a hydrotransport pipeline prior to aeration. The hydrophilic-to-hydrophobic

transition is induced by the diffusion of maltene through the asphaltene layer. Most of the

diffusion-controlled processes are slow. One way to expedite the diffusion is to reduce the

viscosity of maltene, which can be accomplished by increasing temperature. It appears, therefore,

61

that the role of temperature in bitumen aeration may be to decrease the viscosity of maltene. That

contact angle data obtained at different temperatures and presented in Figures 2.9 and 2.10 may

support this view.

After the bitumen-bubble attachment has occurred, bitumen would spread spontaneously

over an air bubble (Figure 2.12b and 2,12c) if the Gibbs free energy change associated in spreading

is negative (Gs < 0), as discussed previously. (Gs) can be calculated as

wwbbsG / [2.10]

Eq. [2.10] is essentially the same as Eq. [2.4]. Contrary to bitumen-air attachment, low

bitumen/water interfacial tension ( wb / ) will favor the bitumen spreading over air bubbles.

Figure 2.12c shows that an air bubble enters a bitumen droplet and remains spherical. The

air bubble may form a coaxial interlayer between the maltene core and the layer of bitumen

exposed on the surface. According to Eq. [2.4], however, sG will become more negative if o

is small, which can be achieved if the bubble remains spherical so that the air/oil (or air/maltene)

interfacial area remains small.

Eqs. [2.9] and [2.10] give the following relation,

wbsa GG /2 [2.11]

which shows that Ga is always more negative than or equal to Gs. Based on Eq. [2.11], there

are three possibilities, representing three possible cases in bitumen aeration. First, if Ga is positive,

Gs has to be positive too. In this case, bitumen will neither attach nor spread on air bubble surfaces

and therefore flotation won’t occur. Second, if Ga is negative, but Gs is positive, in which case

bitumen will attach but won’t spread on air bubbles. This case will occur when the processing

temperature of the bitumen extraction is low, e.g., < 35 °C, and bitumen droplets will attach to air

bubbles as discrete particles.30, 31 In this case, the success of flotation depends on whether the slurry

62

is sufficiently quiescent that the attached bitumen won’t be sheared away from the bubble. Finally,

if both Ga and Gs are negative, then bitumen will attach and engulf air bubbles, which happens

at high processing temperatures, e.g., > 45 °C. Once the bitumen engulfs an air bubble, only very

high mechanical shear would cause it to be stripped away. Therefore, this is the best condition for

successful bitumen aeration.31

Our thermodynamic analysis shows that bitumen/water interfacial tension ( wb / ) plays an

important role in bitumen extraction form oil sands. Low bitumen/water interfacial tension will be

favorable for bitumen liberation from oil sands as well as the engulfment of bitumen over air

bubbles. On the other hand, high bitumen/water interfacial tension is beneficial for the attachment

between bitumen and air bubble. Therefore, our analysis suggests that an optimum bitumen/water

interfacial tension is needed for bitumen extraction in industrial operations. Bitumen/water

interfacial tension could be controlled by the slurry pH. A high pH will facilitate the ionization of

polar molecules at the bitumen/water interface and therefore reduce the bitumen/water interfacial

tension. It was reported that an optimum amount of NaOH was needed to achieve a maximum

bitumen recovery.32 An over-dose of NaOH would cause a too low bitumen/water interfacial

tension, making bitumen-bubble attachment difficult. It could also cause bitumen to emulsify,

leading to small bitumen droplets that are undesirable for bitumen flotation.33

2.4 Summary and Conclusion

The wetting films of water formed on bitumen and its components (maltene and asphaltene)

thin faster with increasing temperature, which provides an explanation for the improved bitumen

extraction at higher temperatures. At a given temperature, the kinetics increase in the order of

asphaltene, bitumen and maltene.

63

From the experimental data obtained in the film thinning kinetics studies, the disjoining

pressures in the wetting films of water formed on bitumen, maltene, and asphaltene have been

determined. The results show that < 0 for maltene and bitumen, while > 0 for asphaltene. In

general, disjoining pressure decreases with temperature with all three substrates. That the

disjoining pressure for bitumen decreases with temperature provides an explanation for the

increased kinetics of bubble-bitumen interaction and hence improved extraction at higher

temperatures.

The dynamic contact angle measurements conducted on bitumen show that, initially,

contact angles are small but subsequently increase above 90o, obviously due to changes in the

interfacial tensions involved. It appears that the changes in interfacial tensions are brought about

by the diffusion of maltene from underneath the asphaltene layer formed on the surface of bitumen,

causing an increase in the surface hydrophobicity of bitumen.

References

1. Helper, L.G. and C. Hsi, AOSTRA Technical Handbook on Oil Sands, Bitumen and Heavy

Oils, AOSTRA technical publication series 6, Alberta Oil Sands Technology and Research

Authority, Edmonton, AB, 1989.

2. Luo, P. and Y. Gu, Effects of asphaltene content on the heavy oil viscosity at different

temperatures, Fuel. 86(7), p. 1069-1078, 2007.

3. Plitt, L.R., Athabasca Tar Sands, in Milling Practice in Canada, p. 371-378, Harpell's

Press Cooperative, Ste. Anne de Bellevue, Quebec, 1978.

64

4. Caudwell, D., et al., The viscosity and density of n-dodecane and n-octadecane at pressures

up to 200 MPa and temperatures up to 473 K, International Journal of Thermophysics.

25(5), p. 1339-1352, 2004.

5. Zhao, H., et al., Effect of divalent cations and surfactants on silica-bitumen interactions,

Industrial & engineering chemistry research. 45(22), p. 7482-7490, 2006.

6. Tanaka, Y., et al., Viscosity and density of binary mixtures of cyclohexane with n-octane,

n-dodecane, and n-hexadecane under high pressures, International journal of

thermophysics. 12(2), p. 245-264, 1991.

7. Clark, K. and D. Pasternack, Hot Water Seperation of Bitumen from Alberta Bituminous

Sand, Industrial & Engineering Chemistry. 24(12), p. 1410-1416, 1932.

8. Evans, E., The viscosity of hydrocarbons. Parts VII and VIII, J. Inst. Pet. Technol. 24 p.

537, 1938.

9. Sury, K.N., Low Temperature Bitumen Recovery Process, patent US 4946597, August 7,

1990.

10. Mankowski, P., et al., Syncrude's Low Energy Extraction Process: Commercial

Implementation, in Proceedings of the 31st Annual Meeting of the Canadian Mineral

Processors, Canadian Mineral Processors, CIM, Ottawa, Canada, 1999.

11. Yoon, R.H., D. Guzonas, and B.S. Aksoy, Role of Surface Forces in Tar Sand Processing,

in Proceeding of the 1st UBC-McGill Bi-Annual International Symposium Vancouver, BC,

Canada, 1995.

12. Gu, G., et al., Effects of physical environment on induction time of air–bitumen attachment,

International Journal of Mineral Processing. 69(1), p. 235-250, 2003.

65

13. Yoon, R.H. and J.L. Yordan, Induction time measurements for the quartz—amine flotation

system, Journal of colloid and interface science. 141(2), p. 374-383, 1991.

14. Scheludko, A. and D. Exerowa, Kolloid-Z. 165 p. 148, 1959.

15. Scheludko, A., Dokl. Bolg. Akad. Nauk. 24 p. 47, 1971.

16. Pan, L., S. Jung, and R.H. Yoon, Effect of hydrophobicity on the stability of the wetting

films of water formed on gold surfaces, Journal of colloid and interface science. 361(1), p.

321-330, 2011.

17. Gu, G., et al., A novel experimental technique to study single bubble–bitumen attachment

in flotation, International Journal of Mineral Processing. 74(1), p. 15-29, 2004.

18. Yoon, R.H. and Y.I. Rabinovich, Role of Asphaltene in the Processing of Tar Sand, in

Proceeding of the 3rd UBC-McGill Bi-Annual International Symposium, CIM, Montreal,

Quebec, Canada, 1999.

19. Ren, S., et al., Effect of weathering on surface characteristics of solids and bitumen from

oil sands, Energy & Fuels. 23(1), p. 334-341, 2008.

20. Long, J., et al., Effect of Operating Temperature on Water‐Based Oil Sands Processing,

The Canadian Journal of Chemical Engineering. 85(5), p. 726-738, 2007.

21. Good, R.J. and L. Girifalco, A theory for estimation of surface and interfacial energies. III.

Estimation of surface energies of solids from contact angle data, The Journal of Physical

Chemistry. 64(5), p. 561-565, 1960.

22. Van Oss, C.J., Interfacial Forces in Aqueous Media, 2nd ed, CRC press, Taylor and Francis

Group 2006.

23. Moran, K., J. Masliyah, and A. Yeung, Factors affecting the aeration of small bitumen

droplets, The Canadian Journal of Chemical Engineering. 78(4), p. 625-634, 2000.

66

24. Lesueur, D., The colloidal structure of bitumen: Consequences on the rheology and on the

mechanisms of bitumen modification, Advances in colloid and Interface Science. 145(1),

p. 42-82, 2009.

25. Kokal, S., et al., Electrokinetic and adsorption properties of asphaltenes, Colloids and

Surfaces A: Physicochemical and Engineering Aspects. 94(2), p. 253-265, 1995.

26. Barbour, R.V. and J.C. Petersen, Molecular interactions of asphalt. Infrared study of the

hydrogen-bonding basicity of asphalt, Analytical Chemistry. 46(2), p. 273-277, 1974.

27. Boduszynski, M.M., J.F. McKay, and D.R. Latham, Proc. Assoc. Asphalt Paving Technol.

49 p. 123, 1980.

28. Moschopedis, S.E. and J.G. Speight, Investigation of hydrogen bonding by oxygen

functions in Athabasca bitumen, Fuel. 55(3), p. 187-192, 1976.

29. Petersen, J.C., Fuel. 46 p. 295, 1967.

30. Masliyah, J., et al., Understanding Water‐Based Bitumen Extraction from Athabasca Oil

Sands, The Canadian Journal of Chemical Engineering. 82(4), p. 628-654, 2004.

31. Schramm, L.L., E.N. Stasiuk, and D. Turner, The influence of interfacial tension in the

recovery of bitumen by water-based conditioning and flotation of Athabasca oil sands, Fuel

Processing Technology. 80(2), p. 101-118, 2003.

32. Sanford, E.C., Processibility of athabasca oil sand: Interrelationship between oil sand fine

solids, process aids, mechanical energy and oil sand age after mining, The Canadian

Journal of Chemical Engineering. 61(4), p. 554-567, 1983.

33. Liu, J., Z. Xu, and J. Masliyah, Colloidal forces between bitumen surfaces in aqueous

solutions measured with atomic force microscope, Colloids and Surfaces A:

Physicochemical and Engineering Aspects. 260(1), p. 217-228, 2005.

67

Chapter 3

Colloidal Forces in Bitumen Aeration

Abstract

The thin film pressure balance (TFPB) technique has been used to study the thinning

kinetics of the wetting films formed on bitumen, asphaltene, and maltene and to determine the

disjoining pressures () in the films from the curvature changes recorded during the process of

film thinning. It was found that film thinning kinetics decreased in the order of maltene, bitumen

and asphaltene. The results also showed that < 0 on maltene and bitumen, while > 0 on

asphaltene, indicating that asphaltene may serve as a kinetic barrier for bitumen aeration. The

disjoining pressure data obtained in the present work have been analyzed in view of the extended

DLVO theory. The results suggest that hydrophobic force is the driving force for bitumen aeration.

3.1 Introduction

In the water-based bitumen extraction from Athabasca oil sands, bitumen aeration is

critically important. Without aeration bitumen would not float because it has almost the same

density as water.1 Bitumen aeration is one of the two fundamental steps in the extraction process,

the other being bitumen liberation from sand grains. These two steps entails three-phase

interactions involving bitumen-bitumen, bitumen-solids (silica and clays), and bitumen-bubble. It

has been generally recognized that colloidal forces play a crucial role in controlling these

interactions and thus the stability of the thin liquid films (TLFs) formed between the three

macroscopic phases.

68

In the past two decades, several researchers have reported their investigations of colloidal

forces between two bitumen surfaces.2-8 They studied various factors affecting colloidal forces and

some also investigated adhesion forces. In this work, we focus on the nature of colloidal forces.

Yoon et. al.2, 3 reported the first direct measurement of colloidal forces between two bitumen

surfaces. They prepared bitumen-coated mica or silica surfaces by means of a Langmuir-Blodgett

trough (L–B deposition) and conducted surface force measurements using SFA and AFM. They

observed long-range repulsive electrostatic force at a separation distance above 70 nm and strong

repulsive force at a shorter range that cannot be explained by the classical DLVO theory. The

authors suggested that the non-DLVO repulsive forces were steric forces due to the asphaltenes

exposed on the bitumen surface in an aqueous environment. Wu et al.4-6 on the other hand,

investigated the colloidal forces between two micron-sized bitumen droplets in water based on the

collision trajectories of two colliding bitumen droplets and the hydrodynamic force applied to

break up a bitumen doublet into two droplets. They suggested that the enhanced repulsion between

bitumen was due to the heterogeneous protrusion structure on the bitumen surface. Laroche et al.7

used the same techniques and compared colloidal forces between deasphalted bitumen (i.e.

maltene) droplets and whole bitumen droplets and found that the repulsion increases in the

presence of asphaltene. More recently, Liu et al.8 measured colloidal forces between two bitumen

surfaces also using AFM as Yoon did. The difference was that Liu et al. prepared the bitumen

substrate using spin-coating method and the bitumen sphere using dip-coating technique. In

addition to the repulsive long-range electrostatic force and short-range steric force, they also

detected attractive hydrophobic force at a separation distance around 10 nm in the presence of

1mM KCl in the solution.

69

For bitumen-sand interaction, Liu et al.9 were the first to directly measure the surface force

between bitumen and silica with AFM and found that at long separation distances bitumen-silica

repulsion could be described by DLVO theory while at shorter distance (< 2–3 nm) an additional

repulsive force of steric origin was observed. Long et. al.10 found that the repulsive force between

bitumen and silica increases with increasing temperature. Later, Zhao et al.11 measured the surface

force between bitumen and silica also using AFM in synthetic plant solutions, actual plant water,

and the foam and residue solutions. In the synthetic electrolyte solutions, they found similar results

as Liu et al.9. While in plant process water and the foam and residue solutions, the measured forces

were weaker than the DLVO forces. More recently, Hogshead et al.12 used AFM to measure

bitumen-silica interaction forces in both an aqueous solution and an ionic liquid. When Utah

bitumen was used, colloidal forces collected in water and ionic liquid appeared similar. For the

Alberta bitumen, the attractive force was longer-ranged and stronger in water than in the ionic

liquid. For bitumen-clay interactions, Liu et al.13 measured the surface forces between bitumen-

kaolinite and bitumen-montmorillonite using AFM. Repulsive forces were detected for both

systems.

Many researchers studied the bitumen-bubble interactions.2, 14-24 As far as colloidal forces

are concerned, only a couple papers have been published in the open literature. In these studies,22,

23 the forces between an air bubble and a bitumen surface were measured with AFM, in which the

air bubble was placed on a hydrophobic substrate, while bitumen was coated on silica spheres by

dip-coating. Long range repulsive forces were initially observed due to electrostatic double layer

forces. It was also reported that an external force applied to the AFM cantilever was needed to

exceed a threshold value before the probe jumped into contact with the air bubble, forming stable

three-phase contact line.

70

Studies of the wetting film formed between bitumen and air bubble can provide useful

information for better understanding the colloidal forces acting between the two macroscopic

surfaces. In 1967, Laskowski and Kitchener25 recognized that the disjoining pressure () in

wetting films must be negative for bubble-particle attachment to occur in flotation. However, the

authors recognized the difficulty in measuring the negative disjoining pressures because the

wetting films with < 0 are unstable and rupture too quickly. It was only recently that CAST

developed a technique to measure negative disjoining pressures using a modified thin film pressure

balance (TFPB).26, 27 This new method is based on monitoring the profiles of the wetting films in

nanoscale and analyzing the profiles using the Reynolds lubrication theory to determine the excess

pressure in the film. This was made possible by monitoring the deformation of bubbles during the

course of bubble-particle interactions.

Bitumen is a colloidal dispersion of asphaltene micelles in maltenes.28, 29 Asphaltenes are

the insoluble part of a bitumen in low molecular weight paraffins (e.g., heptane, pentane, and

hexane) but soluble in light aromatic hydrocarbons (e.g., toluene and benzene). Maltenes are the

soluble part in both types of the solvents and consists of saturates, aromatics and resins. The polar

components of a bitumen are asphaltenes and resins, but asphaltenes are more polar than resins.

These polar molecules have strong affinity toward water and adsorb at oil/water interfaces,

contributing to the stabilization of the emulsions formed in bitumen extraction process.30 For the

bitumen-in-water emulsions, some researchers31-33 suggested that they are stabilized by natural

surfactants, mainly carboxylates and sulfates, present in bitumen, while others2, 3 proposed that

asphaltenes are the stabilizers. For the water-in-bitumen emulsions, both asphaltenes and resins

are responsible for the stabilization of the emulsions, but asphaltenes are the key stabilizer.34-37 In

these regards, one may expect that the asphaltenes and/or resins may play a significant role. It is,

71

therefore, the objective of this chapter to study the role and extent that asphaltenes play in bitumen

aeration.

In the present study, asphaltenes and maltenes were separated from bitumen to determine

the colloidal forces present in the wetting films of water formed on these substrates and to discuss

their roles in bitumen aeration. The kinetics of thinning of the wetting films formed on bitumen,

asphaltene, and maltene were monitored using the TFPB technique. The disjoining pressure was

determined based on the temporal and spatial profiles of the wetting films by the method developed

by Lei et al.27 The disjoining pressure data obtained in the present work have been analyzed in the

view of the extended DLVO theory.3, 38

3.2 Experimental

3.2.1 Materials

A sample of bitumen (99 wt%) was provided by Suncor Energy, Inc. Reagent grade toluene

(>99.5% purity) was purchased from Spectrum Chemical Mfg. Corp., and HPLC grade n-heptane

(99.3% purity) was obtained from Fisher Scientific. Millipore ultrapure water with a resistivity of

18.2 MΩ·cm was obtained using a Direct Q3 water purification system. Silicon wafers (100 crystal

planes, University Wafer) were used as the substrates for coating bitumen, asphaltene and maltene.

Sulfuric acid (H2SO4, 98% purity, VMR international) and hydrogen peroxide (H2O2, 29.0–32.0%

purity, Alfa Aesar) were used to clean the silicon substrates.

3.2.2 Separation of Asphaltene and Maltene

The bitumen sample was dissolved in toluene at a toluene/bitumen ratio of 5:1 by volume.

The solution was then stirred using a magnetic stirrer for 2 hr, centrifuged for 30 min to remove

the solids, and subsequently put under a fume hood to allow the toluene to naturally evaporate.

After the toluene evaporation was complete, n-heptane was added to the toluene- and solids-free

72

bitumen at a heptane/bitumen ratio of 40:1 by volume. The n-heptane and bitumen mixture were

stirred by means of a magnetic stirrer for 2 hr, during which time maltene dissolved into n-heptane

while asphaltene precipitated out. The mixture was left overnight for the asphaltene to settle at the

bottom. The supernatant n-heptane was carefully removed by means of a pipette and placed under

a fume hood to let the n-heptane naturally evaporate, leaving maltene behind. The amount of the

evaporated n-heptane was small.

The asphaltene precipitate was repeatedly washed for 17 times with an excess amount (~1L)

of n-heptane to remove any maltene that may have co-precipitated. In each washing step, the

mixture was agitated for 2 hr and left to stand overnight to allow the asphaltene to precipitate and

the supernatant was carefully removed. After the last washing step, the supernatant n-heptane

became colorless. The asphaltene precipitate was left under a fume hood to let n-heptane to

evaporate and thereby obtain a sample of purified asphaltene. The amount of the evaporated n-

heptane was small.

3.2.3 Preparation of Bitumen-, Maltene-, and Asphaltene-coated Surfaces

Silicon wafers were first oxidized in an oxidation furnace at 1070 °C for ~40 min using

wet oxidation procedure. The oxidized silicon wafers were then cut into 12x12 mm pieces and

used as substrates for bitumen, asphaltene, and maltene. Before the coating, the silicon substrates

were subjected to ultrasonic vibration in water for ~1 hr and then cleaned by immersing in a

Piranha solution (3:7 by volume of H2O2/H2SO4) at 120 °C for 1 hr, followed by rinsing with an

adequate amount of ultrapure water and pure nitrogen (N2) blow-drying. The silicon substrate was

placed in a spin-coater (Laurell, WS-400-6NPP), rinsed with toluene, and then spun at 6,000 rpm

for 30 s to dry off the toluene. When the substrate was free of toluene, a small amount of (2 or 3

drops) of a bitumen-in-toluene solution (2.5 mg/mL) was placed on the substrate and let it stand

73

for 10 s. The substrate was then spun at 3,500 rpm for 20 s before placing another drop of the

bitumen solution and continuing spinning for additional 10 s. The spinning continued for an

additional 1 minute at 6,000 rpm to evaporate the solvent and obtain a smooth and uniform bitumen

coating. The spin-coated bitumen was then placed in a particulate matter-free laminar hood for

~30 min to ensure maximum removal of toluene from the coating. The procedure described above

was also employed for the preparation of asphaltene- and maltene-coated silicon surfaces.

3.2.4 Measurement of Film Thinning Kinetics

Figure 3.1 shows the modified thin film pressure balance (TFPB) apparatus used in the

present work to study the kinetics of film thinning, which is the same as Figure 2.1. The original

TFPB apparatus was designed for the study of foam films (or the water films between two air

bubbles) by Scheludko and Exorowa.39 The equipment was modified to study the kinetics of film

Figure 3.1 Schematics of the modified TFBC apparatus used for the study of wetting

films.

74

thinning for wetting films.40 As shown in the inset, a flat surface of interest is placed on the top of

a film holder (2.0 mm diameter glass tubing) with its surface facing downward, so that the surface

is in contact with the water in the film holder. In the present work, the flat silicon wafers coated

with bitumen, asphaltene, or maltene were used as samples of interest.

In a given experiment, the silicon wafer-glass tubing assembly was inserted in a glass cell

surrounded by a water jacket, in which water of known temperature was circulated. The film

holder-water jacket assembly was placed on an inverted microscopic stage (Olympus IX51) to

monitor the changes in film thicknesses as a function of time. A mercury arc short lamp (100 W,

Olympus) was used as a light source with a bandpass interference filter (#65-643, Edmund Optics)

to produce a monochromatic green light source (λ=546 nm). In a given experiment, the liquid in

the film holder is slowly withdrawn by turning the knob of the piston. Once the interference

patterns (Newton rings) began to appear in the microscopic field of view, the film was allowed to

thin spontaneously while recording the images by means of a high-speed camera (Fastec Imaging,

HS400115). The interference patterns recorded were used to reconstruct timed thickness profiles

of a wetting film across the entire film holder with a nanometer resolution. The spatial and

temporal film profiles of the wetting film were then used to monitor the changes in film thicknesses

(h) as a function of time and radial distance of the film at any point of the film. The experimental

data were used to obtain information on the film thinning kinetics and calculate the disjoining

pressure () in the thin liquid films (TLFs) according to Eq. [3.1], as derived by Pan. et.al.,27

r

r

r

rdrdr

t

hr

rhr

hr

rrR 03

112

2

[3.1]

where is the air/water interfacial tension, R the bubble radius, r the radial position, h the wetting

film thickness, and µ the fluid viscosity.

75

3.3 Results and Discussion

3.3.1 Film Thinning Kinetics and Disjoining Pressure

Figure 3.2 shows the spatial and temporal profiles of the wetting films formed between an

-40 -20 0 20 400

200

400

600

h (

nm

)

(c)

(b)

h (

nm

)h (

nm

)

0.6

0.4

0.2

0 s

Maltene(a)

-40 -20 0 20 400

200

400

600

4.44

3

1

0 s

Bitumen

-40 -20 0 20 400

300

600

r (m)

12.05

4

1

0 s

Asphaltene

Figure 3.2 Spatial and temporal profiles of the wetting films of water formed on maltene (a),

asphaltene (b), and bitumen (c) at 55 °C. The arrows represent the critical thicknesses (hcr) where

the films rupture.

76

air bubble and maltene, bitumen, and asphaltene surfaces in water at 55 °C. The profiles are similar

to those obtained at 80 °C (see Figure 2.3 in Chapter 2). From a thickness of 300 nm, the films

formed on maltene thinned fastest and ruptured at 0.6 s. On asphaltene, the film thinned 20 times

slower than on maltene, with the rupture occurring at 12.05 s. On bitumen, which contains both

maltene and asphaltene, the film thinned slower than on maltene and faster than on asphaltene.

The film thickness (h) vs. time (t) plots in Figure 3.3 presents the kinetics of film thinning at the

center of the wetting film, which increased in the order of asphaltene, bitumen, and maltene. It is

obvious from Figure 3.3 that the film on asphaltene seemed “metastable” for about 3 s before its

rupture, corresponding to the flattening of the film shown in Figure 3.2.

The film profiles presented in Figure 3.2 have been used to obtain kinetic information such

as 𝜕ℎ 𝜕𝑡⁄ and 𝜕ℎ 𝜕𝑟⁄ (r is the radius position) that can be used to calculate the disjoining pressure

() according Eq. [3.1].27

0 4 8 120

100

200

300

t (s)

MalteneBitumen

Natural pH

h (

nm

)

55 0C

Asphaltene

Figure 3.3 Kinetics of film thinning on maltene, bitumen, and asphaltene at 55 °C. It decreases

in the order of maltene, bitumen and asphaltene.

77

Figure 3.4 shows the results of obtained for maltene, bitumen, and asphaltene at different

time and radial positions. In these calculations, 67.1 mN/m, µ = 0.5x10-3 Pas at 55 °C, and R

= 2 mm were used. As shown, the disjoining pressures were negative in the wetting films formed

on both maltene and bitumen, i.e., < 0, while it is positive, i.e., > 0, on asphaltene. A vertical

pressure balance shows that

curpp [3.2]

where p is the pressure of the liquid in the wetting film and curp is the pressure due to curvature

change and surface tension (Laplace pressure). Eq. [3.2] shows that a negative disjoining pressure

helps increase the film thinning process, while a positive disjoining pressure causes retardation.

Thus, the results presented in Figure 3.4 explain why the kinetics of film thinning increased in the

order of asphaltene, bitumen, and maltene. The change in disjoining pressure with film thickness

at the center of the films on maltene, bitumen, and asphaltene were shown in Figure 3.5. As shown,

on maltene and bitumen, the disjoining pressure becomes increasingly negative as the film

thickness decreases, while on asphaltene it becomes increasingly positive. = -450 Pa on maltene

and = -160 Pa on bitumen were obtained just before the wetting films rupture. Therefore, the

fast kinetics observed with maltene can be attributed to the highly negative disjoining pressure.

It has been shown that the hydrophobic force, and hence the magnitude of negative

disjoining pressure, increase with increasing surface hydrophobicity.41 The results obtained in the

present work show, therefore, that maltene is the most hydrophobic, followed by bitumen.

Asphaltene, on the other hand, is hydrophilic as shown by the positive disjoining pressures

measured (see Figures 3.4b and 3.5). Asphaltene being least hydrophobic is understandable

because its functional groups, such as carboxylic, amide, thiol, hydroxyl, and carbonyl, are

78

-400

-200

0

(c)

(b)

(

Pa)

(

Pa)

0.6

0.4

0.2

0 s

r (m)

(

Pa)

Maltene

0

30

60

90

0

2

6

12.05 s

0 5 10 15 20-200

-100

0

Asphaltene

Bitumen

4.44

3

0 s

0 5 10 15 20

(a)

Figure 3.4 Changes in the disjoining pressure () of the wetting films of water formed on

maltene (a), asphaltene (b), and bitumen(c) as functions of time and radial distance (r) of the

film from the film center at 55 °C.

79

hydrophilic. Maltene being the most hydrophobic is also understandable, because it contains

hydrocarbons. The hydrophobicity of bitumen lies in between asphaltene and maltene, suggesting

that asphaltene is exposed to the bitumen/water interface in aqueous.

The results obtained in the present work suggest that the hydrophobicity of maltene is

compromised by the presence of asphaltene on the surface. It is likely that the hydrophilic groups

of asphaltene are exposed to the aqueous phase, the extent of which will determine the

hydrophobicity of bitumen and its floatability. That bitumen is still hydrophobic despite the likely

exposure of asphaltene on the surface suggests that the surface may not be completely covered by

the hydrophilic groups of asphaltene, which is very likely since more than 80% of bitumen is

maltene. Further study is necessary to more fully understand the hydrophobicity of bitumen.

0 100 200 300

-400

-200

0

Natural pH

Asphaltene

Bitumen

h (nm)

Maltene

55 0C

(

Pa)

Figure 3.5 Changes in the disjoining pressure () at the center of the wetting films of water

formed on maltene, bitumen, and asphaltene at 55 °C. Thermodynamically, the film can rupture

when < 0.

80

3.3.2 Extended DLVO Theory

It is well known that colloidal forces play an important role in bitumen extraction. The

classical DLVO theory42, 43 recognizes two colloidal (or surface) forces, which include the

electrical double-layer force (Fe) and the van der Waals dispersion force (Fd). One can readily

convert the forces to disjoining pressures () or vice versa from the Derjaguin approximation,44

)(2 hGR

F [3.3]

and the definition of disjoining pressure,

sTPhGh ,,)/()( [3.4]

in which F is the surface force, R is the radius of curvature of the macroscopic surface(s) involved,

and G(h) is the free energy of a wetting film which varies with film thickness h or the separation

between two surfaces. According to the DLVO theory, the disjoining pressure of a thin liquid film

(TLF) can be given by,

de [3.5]

where the subscripts e and d represent the electrical double-layer and van der Waals dispersion

forces, respectively.

The DLVO theory has long been used as the governing law in a wide range of industries,

including the bitumen industry. The theory can predict the stability of the thin liquid films (TLFs)

formed between any two surfaces involving macroscopic particles, air bubbles, oils, etc. For

example, the disjoining pressure in the TLF between bitumen and an air bubble affects the bitumen

aeration during the hydrotransport process.

When a particle approaches an air bubble, the TLF (or wetting film) formed in between

them must be thinned quickly and rupture (or break) to form a finite contact angle ().

81

Thermodynamically, wetting films rupture when > 0, as is well known. The role of colloidal

forces in bubble-particle interactions can be better appreciated using the Frumkin-Derjaguin

isotherm,45

0

)/1(1cos 23h

dh [3.6]

in which 23 is the interfacial tension between air bubble and water (or surface tension of water),

and is the disjoining pressure in a wetting film which varies with film thickness h. The integral

of from h = ho to represents the Gibbs free energy change associated with bubble-particle

interaction (G). Here, the parameter ho represents the thickness of the water film in equilibrium

with the vapor phase inside an air bubble attached to a hydrophobic surface. The values of ho are

usually less than 1 nm.

The disjoining pressure in the wetting film due to the van der Waals force is given by,

3

132

6 h

Ad

[3.7]

where 132A is the Hamaker constant, while the disjoining pressure due to the electrostatic double-

layer force is given by

)coth(2)(cos()sinh(2

21

2

2

2

1

2

0 hhechh

e

[3.8]

in which 0 is the permittivity in vacuum, the dielectric constant of water, 1 and 2 the

double-layer potentials at the solid/water and air/water interfaces, respectively, and is the

reciprocal Debye length. In this report, the subscripts 1, 2, and 3 represent solid, gas, and water

phases, respectively.

Substituting Eqs. [3.7] and [3.8] into the DLVO theory (Eq. [3.5]) and subsequently into

the Frumkin-Derjaguin isotherm (Eq. [3.6]), one obtains the following equation,

82

1)2exp(

)exp(2

12

11cos

0

2

2

2

1021

02

0

132

23

0h

h

h

A

[3.9]

In wetting films, both of the terms within the bracket are positive values, because A132 is negative

and ψ1 and ψ2 are negative under most flotation conditions including bitumen aeration. Under these

conditions, contact angles cannot be greater than zero and consequently G > 0 for the bitumen-

bubble interaction. The only way to have > 0 and G < 0, would be to create a hydrophobic force

in wetting films, so that one can use the following relation,

)exp(

21-)2exp(

--)exp(2

12-

11cos

0

2

2

2

1021

2

0

132

23 D

hC

h

h

h

A oo

[3.10]

in which the third term in the bracket represents the contribution from the hydrophobic force (h)

to the total disjoining pressure () in a wetting film.

According to Eq. [3.10], the hydrophobic force (and hence the negative disjoining pressure)

must be large enough to overcome the sum of the repulsive van der Waals and double-layer forces

in order to break the wetting film and form a finite contact angle that is greater than zero. The three

terms within the bracket constitute the extended DLVO theory,46

hde [3.11]

where h represents the disjoining pressure due to the hydrophobic force in a TLF, as proposed

earlier by Yoon’s group at Virginia Tech.26, 27

3.3.3 Disjoining Pressure Isotherms

According to the extended DLVO theory (Eq. [3.11]), the disjoining pressure () in a

wetting film has three components, i.e., double-layer (e), van der Waals dispersion (d), and

hydrophobic force (h) components. It will be of interest to construct disjoining pressure

isotherms, in which changes in the disjoining pressure components vs. film thickness are plotted.

83

a) Maltene

In Figure 3.6, the changes in e, d, and h vs. h are plotted for the wetting film of water

formed on maltene at 55 °C. The triangles represent the experimental data obtained in the present

work, and the solid line represents a fit to the extended DLVO theory (Eq. [3.11]). There are three

dashed lines, representing d calculated using Eq. [3.7] with a A132 value of -1 x 10-20 J, and e

calculated using Eq.[3.8] with ψ1 = - 65 mV for the surface potential of maltene, and ψ2 = - 26 mV

for that of air bubble, and κ-1 = 50 nm for the Debye length, and

D

h

D

Ch exp

2 [3.12]

Figure 3.6 The disjoining pressure isotherm ((h)) obtained at the center of the wetting film

of water formed on maltene at 55 °C. The triangles represent the experimental data presented

in Figure 3.5. The solid line represents a fit to the extended DLVO theory, which includes

contributions from: van der Waals force (d) with a Hamaker constant of A132 = -1 x 10-20 J;

electrostatic force (e) with surface potential of ψ1 = - 65 mV for maltene, surface potential

of ψ2 = - 26 mV for air bubble, and Debye length of κ-1 = 50 nm; and hydrophobic force (h)

with fitting parameters of C = -0.43 mN/m and D = 50 nm.

84

where C (= -0.4 mN/m) and D (= 50 nm), known as decay lengths, are the best fit parameters. The

figure legend shows all other parameters used to construct the disjoining pressure isotherms.

The vs. h isotherms show that h far exceeds in magnitude the sum of e and d, so that

is net negative at all separations studied. The decay length of 50 nm is very large as compared

to the hydrophobic forces measured between hydrophobic surfaces, partly because air bubbles in

water are the most hydrophobic entity known to date.47, 48 Air bubbles in water have an interfacial

tension of 72 mN/m, while hydrocarbon oils have interfacial tensions of 50 mN/m. Interfacial

tensions are excellent measures of hydrophobicity.

b) Asphaltene

Figure 3.7 shows the disjoining pressure isotherms obtained at the center of the wetting

film of water formed on asphaltene at 55 °C. As shown, > 0 indicating that asphaltene is

hydrophilic. As suggested previously,2 the sharp increase in may represent the repulsive steric

disjoining pressure (s). Thus, the extended DLVO theory may be written as follows,

sed [3.13]

where

4/39/4

3)2/()/2( LhhL

s

kTs [3.14]

In Eq. [3.14], s represents the mean distance between grafted points of a polymer, and L is the

length of the polymer tail. Eq. [3.14] was derived by de Gennes for the interaction between two

85

surfaces grafted by “brush-like” polymers.49 The best-fit parameters used to fit the experimental

data are presented in the figure legend.

c) Bitumen

Figure 3.8 shows the disjoining pressure isotherms for the TLF of water formed on bitumen

at 55 °C. The extended DLVO theory that was used to represent the system included four different

components:

hsed [3.15]

Figure 3.7 The disjoining pressure isotherm ((h)) obtained at the center of the wetting film of

water formed on asphaltene at 55 °C. The triangles represent the experimental data presented in

Figure 3.5. The solid line represents a fit to the extended DLVO theory, which includes

contributions from: van der Waals force (d) with a Hamaker constant of A132 = -1 x 10-20 J;

electrostatic force (e) with surface potential of ψ1 = - 68 mV for asphaltene, surface potential

of ψ2 = - 26 mV for air bubble, and Debye length of κ-1 = 65 nm; and repulsive steric force (s)

between asphaltene polymers with a brush length of L = 63 nm and a grafting distance of s = 23

nm.

86

where the subscripts d, e, s, and h represent the dispersion, double-layer, steric, and hydrophobic

disjoining pressures, respectively. The model parameters that were used to fit the experimental

data are given as figure legends.

As shown in Chapter 2, disjoining pressure measurements were also conducted with

bitumen substrates at other three different temperatures, i.e., 22, 35 and 80 °C, and their disjoining

pressure isotherms were constructed using the extended DLVO theories (Eq. [3.15]) with the

Figure 3.8 The disjoining pressure isotherm ((h)) obtained at the center of the wetting film of

water formed on bitumen at 55 °C. The triangles represent the experimental data. The solid line

represents a fit to the extended DLVO theory, which includes contributions from: van der Waals

force (d) with a Hamaker constant of A132 = -1 x 10-20 J; electrostatic force (e) with surface

potential of ψ1 = - 57 mV for bitumen, surface potential of ψ2 = - 30 mV for air bubble, and Debye

length of κ-1 = 45 nm; repulsive steric force s) between bitumen polymers with a brush length

of L = 32 nm and a grafting distance of s = 26 nm, and hydrophobic force (h) with fitting

parameters of C = -0.2 mN/m and D = 54 nm.

87

fitting parameters shown in Table 3.1. The disjoining pressure component due to hydrophobic

force (h) have been obtained and plotted together in Figure 3.9. As shown, h becomes

increasingly negative (or strong) and longer-ranged at higher temperatures, which is consistent

with the flotation data reported in the literature.1 Further, this finding is in agreement with the

contact angle data obtained on the thick films of bitumen immersed in water (Chapter 2, Figures

2.9 and 2.10).

That hydrophobic forces measured in the wetting films of bitumen becomes stronger with

increasing temperature, as shown in Figure 3.9, is contrary to what has been reported previously

by Yoon’s group at Virginia Tech.50, 51 The group conducted a series of AFM force measurements

with silica and gold surfaces hydrophobized by different surfactant coatings. The results obtained

with the gold and silica surfaces hydrophobized with n-hexadenane (C-16) thiol and

octadecylchlorosilane (OTS), respectively, showed that hydrophobic forces increase with

decreasing temperature, which is contrary to what has been observed in the present work with

bitumen and presented in Figure 3.9.

Table 3.1 Parameters used to fit the disjoining pressure of the wetting film of water

formed on bitumen at different temperatures

T A132 ψ1 ψ2 κ-1 s L C D

( °C) ( J ) ( mV ) ( mV ) ( nm ) ( nm ) ( nm ) ( mN/m ) ( nm )

22 -1 x 10-20 -50 -24 80 30 68.3 0 0

35 -1 x 10-20 -55 -20 33 26 28.3 -0.12 47

55 -1 x 10-20 -57 -30 45 26 32 -0.2 54

80 -1 x 10-20 -110 -45 50 26 32 -1.32 55

88

A rigorous thermodynamic analysis of the AFM force data obtained previously by the

Yoon group showed that the excess entropy (Sf) in the thin liquid films (TLFs) confined between

hydrophobic surfaces decreases with decreasing film thickness, which has been attributed to the

formation of H-bonded structures in the TLFs. The free energy of the water molecules in the TLFs

confined between hydrophobic surfaces are higher than those in thick films, because much of the

water molecules in the vicinity of hydrophobic surfaces cannot form H-bonds with the confining

surfaces. The excess free energy is expended to form H-bonded structures, known as low density

liquid (LDL).

It has been found also that the entropy decrease (Sf < 0) becomes more significant at lower

temperatures, which indicates that water molecules can more readily form structures at lower

0 100 200 300

-1500

-1000

-500

0

80 oC

55 oC

h (

Pa)

h (nm)

35 oC

Natual pH

Bitumen

Figure 3.9 Hydrophobic disjoining pressure isotherms (h) on bitumen at different

temperatures in water of a natual pH of about 5.7.

89

temperatures, which is consistent with our daily experiences that water forms H-bonded structures

at lower temperatures (i.e., ice).

It is well known that water molecules form low density liquid (LDL) in supercooled

water.52 More recent spectroscopic studies conducted using x-ray absorption spectroscopy (XAS)

and x-ray Raman scattering (XRS) at Stanford University showed that LDL is also formed in bulk

water at ambient temperatures.53, 54 By monitoring the OH-vibrations of water molecules using the

Raman and FTIR experiments, the Mallanace group at the University of Messina, Italy, showed

that the population of LDL species at ambient temperatures are much higher than anticipated.55, 56

According to Stillinger, water molecules behave similarly under supercooled conditions and in

hydrophobic interactions,52 suggesting that LDL can be more readily formed in the TLFs confined

between hydrophobic surfaces at ambient temperature. In supercooled water, water molecules are

prevented from forming ice by mechanical agitation or other external forces. Under this condition,

they can form LDL to minimize free energy. In confined spaces, water cannot interact with the

hydrophobic surfaces. One way to expand the excess free energy is to form LDL. Thus, the

hydrophobic forces operating in colloid chemistry and mineral flotation originate from the

structural changes of water in TLFs relative to the water structure in the bulk.

As has already been noted, the hydrophobic forces measured with gold and silica surfaces

coated with C-16 thiol and OTS-coated surfaces decreased with increasing temperature, which is

contrary to what has been observed with bitumen-coated surface in the present work. The AFM

force measurements were carried out in narrow temperature ranges, i.e., 10 to 40 °C, so that the

chemistry of the hydrophobic surfaces would not change significantly with temperature. The only

change that can occur over the narrow temperature range would be the water structure, or, rather,

the tendency for low-density-liquid (LDL) to form in confined spaces.50, 51 With bitumen, the

90

structure and properties of the substrate, i.e., bitumen, can change significantly over the larger

temperature range investigated in the present work, i.e., 22 to 80 °C. After all, the bonding between

hydrocarbons constituting bitumen was in liquid form; therefore, its structure can be more readily

changed than the structure of a solid, such as gold or silica. It has been shown that the surface

tension of bitumen decreases significantly with increasing temperature,57 which is a manifestation

of changing intermolecular bond energy. Surface tensions of solids do not change significantly

over the narrow range of temperatures employed in the present work. In general, the lower the

surface energy (or surface tension), the higher the hydrophobicity. Therefore, the increase in

bitumen contact angle with temperature observed in the present work may be attributed to the

decrease in surface free energy of bitumen, which corroborate well with the surface tension data

noted above. Yoon’s group showed also that the higher the contact angle, the stronger and longer-

ranged the hydrophobic force.41 Thus, there are two opposing factors affecting the hydrophobic

forces in a bitumen flotation system. One is the contact angle, and the other is water structure. It

appears that the former plays a more significant role than the latter.

Figure 3.10 compares the disjoining pressure components due to the steric forces present

in the TLFs formed on bitumen at different temperatures. As shown, s is maximum at 55 °C,

followed by those obtained at 35 °C and 22 °C. The data seem to suggest that asphaltene polymers

are more readily exposed at higher temperatures.

In Table 3.1, the Hamaker constant (A132) for the asymmetric interaction between air bubble

1 and bitumen 2 in the medium of water 3 is predicted using the geometric mean combining rule

232131132 AAA [3.16]

where A131 and A232 are the Hamaker constants for symmetric interactions. From the values of A131

= 3.7 × 10-20 J and A232 = 2.8 × 10-21 J reported in the literature,8, 58 one can obtain the value of A132

91

= -1.0 × 10-20 J for the bitumen and bubble interaction, which is the same as used by Englert et

al.22 The -potentials of the bitumen/water interface given in the table were obtained from the data

reported by Dai et al.59 and Liu et al.60 The bitumen zeta potential used at 80 °C is more negative

than those at lower temperatures, which is in agreement with Dai’s findings59 and other materials

such as quartz,61 sodium kaolinite,61, 62 and others.63 Zeta potentials of air bubble in water were

close to the value reported by McShea et al.64 The rest of the parameters in the table were best-fit

parameters obtained on the basis of the experimental disjoining pressure isotherms. Better fit could

have been obtained if the zeta potentials of bitumen and air bubbles were directly measured on the

bitumen samples used in the present work at the relevant temperatures of interest. The zeta

0 50 100 1500

100

200

300

55 oC

35 oC

s (

Pa

)

h (nm)

22 oC

Natual pH

Bitumen

Figure 3.10 Steric disjoining pressure isotherms (s) on bitumen at different

temperatures in water of a natual pH of about 5.7.

92

potentials can be readily measured by electrokinetic methods, and the structural information on

polymers can be measured using the dynamic light scattering, viscometric methods, and the small-

angle neutron scattering (SANS) techniques.

3.3.4 Disjoining Pressure in Liberation

Unlike the oil sands in Utah, the sand grains in Canadian oil sands are thought to be

surrounded by a thin film of water, known as connate water, as shown in Figure 3.11a. The first

step toward extracting bitumen is to create a positive disjoining pressure in the innate water, i.e.,

> 0. According to the extended DLVO theory discussed above, there are two ways to create a

positive disjoining pressure. One is to increase the steric repulsion (s), and the other is to increase

the double-layer repulsion (e). Both of these parameters are optimized by controlling

temperature, pH, and electrolyte concentrations. At an elevated temperature, asphaltene can be

more readily exposed to the surface of bitumen due to decreased viscosity. As shown in Table 2.1,

the viscosity of Athabasca bitumen decreases by nearly three orders of magnitude as temperature

increases from 20 to 75 °C. The presence of asphaltene polymers creates steric repulsion (s) and

Figure 3.11 a) For liberation, it is necessary to create a positive disjoining pressure in the TLFs

of water between bitumen and sand grains, i.e., > 0; for air bubble-bitumen attachment, it is

necessary to create negative disjoining pressures in the TLFs, i.e., < 0.

93

hence a positive disjoining pressure. Further, an increase in pH should ionize the polar groups of

the asphaltene polymers and, thereby, increase the negative surface charge and increase the

repulsive electrostatic disjoining pressure (e). In principle, any electrolyte in process water

should be detrimental to bitumen liberation as it can cause the surface charge to decrease (e.g., by

Ca2+ ion adsorption) and the electrical double-layer to be compressed. However, the double-layer

compression becomes discernable only at very high electrolyte concentrations.

Figure 3.11b shows that a bubble deforms when it encounters a bitumen surface in a

hydrotransport pipeline. The deformation creates an excess pressure (p) in the film, causing the

film to thin. At this point in the film thinning process, the excess pressure is solely dependent upon

the radius R of the bubble and surface tension, i.e., (pcur = 2/R). As the film thinning continues

and reaches a thickness range of 250 to 300 nm, however, disjoining pressure () comes into

play. The film thins further if < 0. If > 0, however, film thinning stops when becomes equal

to the curvature pressure (pcur), because p becomes zero according to Eq. [3.2]. The film is stable

at this equilibrium thickness (he); therefore, the contact angle () becomes zero, and no flotation

will occur. As shown in Eq. [3.10], > 0 only when < 0 by virtue of the hydrophobic force

present in the wetting film. Since the disjoining pressure in the film must be high to liberate

bitumen, the hydrophobic force should be strong and long-ranged enough to create a negative

disjoining pressure, which is a necessary condition for aeration. One way to increase the

hydrophobic force is to increase the temperature for bitumen aeration. In mineral flotation, a

collector is added to render the surface hydrophobic. In bitumen aeration, temperature is raised to

increase contact angle and hence hydrophobic force as has been discussed in foregoing sections.

94

3.4 Summary

In the present work, the bitumen was separated into maltene and asphaltene in order to

study the role of these two components in the interaction between bitumen and air bubble. The

kinetics of thinning of the wetting films formed on bitumen, asphaltene, and maltene were

monitored using the TFPB technique. The results showed that the kinetics decreased in the order

of maltene, bitumen and asphaltene. The disjoining pressure in the wetting film was determined

based upon the temporal and spatial profiles of the wetting films obtained using the modified TFPB

method. It was found that < 0 on maltene and bitumen, while > 0 on asphaltene, indicating

that asphaltene may serve as a kinetic barrier for bitumen aeration. Finally, the disjoining pressure

data obtained in the present work were analyzed in view of the extended DLVO theories.

Comparison of the disjoining pressure isotherms for maltene, bitumen, and asphaltene showed that

the electrostatic and steric forces are responsible for the difficulties that may be encountered in

bitumen-bubble interactions, while the hydrophobic force serves as the driving force for the fruitful

interactions.

References

1. Masliyah, J.H., Z. Xu, and J.A. Czarnecki, Handbook on Theory and Practice of Bitumen

Recovery from Athabasca Oil Sands, Kingsley Publishing Services, 2011.

2. Yoon, R.H., D. Guzonas, and B.S. Aksoy, Role of Surface Forces in Tar Sand

Processing, in Proceeding of the 1st UBC-McGill Bi-Annual International Symposium

Vancouver, BC, Canada, 1995.

95

3. Yoon, R.H. and Y.I. Rabinovich, Role of Asphaltene in the Processing of Tar Sand, in

Proceeding of the 3rd UBC-McGill Bi-Annual International Symposium, CIM, Montreal,

Quebec, Canada, 1999.

4. Wu, X., et al., Interaction Forces between Bitumen Droplets in Water, Langmuir. 15(16),

p. 5244-5250, 1999.

5. Wu, X., et al., Applications of colloidal force measurements using a microcollider

apparatus to oil sand studies, Colloids and Surfaces A: Physicochemical and Engineering

Aspects. 174(1–2), p. 133-146, 2000.

6. Wu, X., T. Dabros, and J. Czarnecki, Determining the Colloidal Forces between Bitumen

Droplets in Water Using the Hydrodynamic Force Balance Technique, Langmuir. 15(25),

p. 8706-8713, 1999.

7. Laroche, I., et al., Dynamic and Static Interactions between Bitumen Droplets in Water,

Journal of Colloid and Interface Science. 250(2), p. 316-326, 2002.

8. Liu, J., Z. Xu, and J. Masliyah, Colloidal forces between bitumen surfaces in aqueous

solutions measured with atomic force microscope, Colloids and Surfaces A:

Physicochemical and Engineering Aspects. 260(1), p. 217-228, 2005.

9. Liu, J., Z. Xu, and J. Masliyah, Studies on bitumen-silica interaction in aqueous solutions

by atomic force microscopy, Langmuir. 19(9), p. 3911-3920, 2003.

10. Long, J., Z. Xu, and J. Masliyah, On the role of temperature in oil sands processing,

Energy & fuels. 19(4), p. 1440-1446, 2005.

11. Zhao, H., et al., Effect of divalent cations and surfactants on silica-bitumen interactions,

Industrial & engineering chemistry research. 45(22), p. 7482-7490, 2006.

96

12. Hogshead, C.G., et al., Studies of bitumen− silica and oil− silica interactions in ionic

liquids, Energy & Fuels. 25(1), p. 293-299, 2010.

13. Liu, J., Z. Xu, and J. Masliyah, Interaction forces in bitumen extraction from oil sands,

Journal of colloid and interface science. 287(2), p. 507-520, 2005.

14. Drelich, J., et al., The role of gas bubbles in bitumen release during oil sand digestion,

Fuel. 74(8), p. 1150-1155, 1995.

15. Drelich, J., D. Leliński, and J.D. Miller, Bitumen spreading and formation of thin

bitumen films at a water surface, Colloids and Surfaces A: Physicochemical and

Engineering Aspects. 116(1), p. 211-223, 1996.

16. Gu, G., et al., Effects of physical environment on induction time of air–bitumen

attachment, International Journal of Mineral Processing. 69(1), p. 235-250, 2003.

17. Gu, G., et al., A novel experimental technique to study single bubble–bitumen attachment

in flotation, International Journal of Mineral Processing. 74(1), p. 15-29, 2004.

18. Wang, W., et al., Attachment of individual particles to a stationary air bubble in model

systems, International Journal of Mineral Processing. 68(1–4), p. 47-69, 2003.

19. Gu, G., et al., Hydrogen and Oxygen Bubble Attachment to a Bitumen Drop, The

Canadian Journal of Chemical Engineering. 82(4), p. 846-849, 2004.

20. Lelinski, D., et al., Rate of bitumen film transfer from a quartz surface to an air bubble as

observed by optical microscopy, The Canadian Journal of Chemical Engineering. 82(4),

p. 794-800, 2004.

21. Moran, K., J. Masliyah, and A. Yeung, Factors affecting the aeration of small bitumen

droplets, The Canadian Journal of Chemical Engineering. 78(4), p. 625-634, 2000.

97

22. Englert, A., et al., Interaction forces between a deformable air bubble and a spherical

particle of tuneable hydrophobicity and surface charge in aqueous solutions, Journal of

colloid and interface science. 379(1), p. 121-129, 2012.

23. Ren, S., J. Masliyah, and Z. Xu, Studying bitumen–bubble interactions using atomic force

microscopy, Colloids and Surfaces A: Physicochemical and Engineering Aspects. 444 p.

165-172, 2014.

24. Su, L., Z. Xu, and J. Masliyah, Role of oily bubbles in enhancing bitumen flotation,

Minerals engineering. 19(6), p. 641-650, 2006.

25. Laskowski, J. and J. Kitchener, The hydrophilic—hydrophobic transition on silica, J. of

Colloid and Interface Sci. 29(4), p. 670-679, 1969.

26. Pan, L., S. Jung, and R.H. Yoon, A fundamental study on the role of collector in the

kinetics of bubble–particle interaction, International Journal of Mineral Processing. 106

p. 37-41, 2012.

27. Pan, L., S. Jung, and R.H. Yoon, Effect of hydrophobicity on the stability of the wetting

films of water formed on gold surfaces, Journal of colloid and interface science. 361(1),

p. 321-330, 2011.

28. Dickie, J.P. and T.F. Yen, Macrostructures of the asphaltic fractions by various

instrumental methods, Analytical chemistry. 39(14), p. 1847-1852, 1967.

29. Lesueur, D., The colloidal structure of bitumen: Consequences on the rheology and on

the mechanisms of bitumen modification, Advances in colloid and Interface Science.

145(1), p. 42-82, 2009.

30. Khadim, M.A. and M.A. Sarbar, Role of asphaltene and resin in oil field emulsions,

Journal of Petroleum Science and Engineering. 23(3–4), p. 213-221, 1999.

98

31. Schramm, L.L. and R.G. Smith, The influence of natural surfactants on interfacial

charges in the hot-water process for recovering bitumen from the athabasca oil sands,

Colloids and Surfaces. 14(1), p. 67-85, 1985.

32. Schramm, L.L. and R.G. Smith, Two classes of anionic surfactants and their significance

in hot water processing of oil sands, The Canadian Journal of Chemical Engineering.

65(5), p. 799-811, 1987.

33. Acevedo, S., X. Gutierrez, and H. Rivas, Bitumen-in-Water Emulsions Stabilized with

Natural Surfactants, Journal of Colloid and Interface Science. 242(1), p. 230-238, 2001.

34. McLean, J.D. and P.K. Kilpatrick, Effects of Asphaltene Solvency on Stability of Water-

in-Crude-Oil Emulsions, Journal of Colloid and Interface Science. 189(2), p. 242-253,

1997.

35. Yan, Z., J.A.W. Elliott, and J.H. Masliyah, Roles of Various Bitumen Components in the

Stability of Water-in-Diluted-Bitumen Emulsions, Journal of Colloid and Interface

Science. 220(2), p. 329-337, 1999.

36. Gao, S., et al., Role of Bitumen Components in Stabilizing Water-in-Diluted Oil

Emulsions, Energy & Fuels. 23(5), p. 2606-2612, 2009.

37. Gu, G., et al., Influence of water-soluble and water-insoluble natural surface active

components on the stability of water-in-toluene-diluted bitumen emulsion, Fuel. 81(14),

p. 1859-1869, 2002.

38. Xu, Z. and R.H. Yoon, A study of hydrophobic coagulation, Journal of Colloid and

Interface Science. 134(2), p. 427-434, 1990.

39. Scheludko, A. and D. Exerowa, Kolloid-Z. 165 p. 148, 1959.

40. Scheludko, A., Dokl. Bolg. Akad. Nauk. 24 p. 47, 1971.

99

41. Yoon, R.H. and S.A. Ravishankar, Long-range hydrophobic forces between mica

surfaces in dodecylammonium chloride solutions in the presence of dodecanol, Journal of

Colloid and Interface Science. 179(2), p. 391-402, 1996.

42. Derjaguin, B., Theory of the stability of strongly charged lyophobic sols and the adhesion

of strongly charged particles in solutions of electrolytes, Acta Physicochim. USSR. 14 p.

633-662, 1941.

43. Verwey, E.J.W. and J.T.G. Overbeek, Theory of the Stability of Lyophobic Colloids,

Elsevier, Amsterdam, 1948.

44. Derjaguin, B.V., Friction and adhesion IV. The theory of adhesion of small particles.,

Kolloid Z. 69 p. 155-164, 1934.

45. Frumkin, A., On the phenomena of wetting and the adhesion of bubbles I, Zh. Fiz. Khim.

12 p. 337, 1938.

46. Yoon, R.H. and L. Mao, Application of Extended DLVO Theory, IV: Derivation of

Flotation Rate Equation from First Principles, Journal of Colloid and Interface Science.

181(2), p. 613-626, 1996.

47. Yoon, R.H. and L. Wang, Hydrophobic Forces in Foam Films, in Colloid Stability- The

Role of Surface Forces - Part I, T.F. Tadros, Editor, p. 161-186, Wiley-VCH, 2006.

48. Van Oss, C.J., Interfacial Forces in Aqueous Media, 2nd ed, CRC press, Taylor and

Francis Group 2006.

49. De Gennes, P., Polymers at an interface; a simplified view, Advances in Colloid and

Interface Science. 27(3), p. 189-209, 1987.

100

50. Wang, J., R.H. Yoon, and J.C. Eriksson, Excess thermodynamic properties of thin water

films confined between hydrophobized gold surfaces, Journal of colloid and interface

science. 364(1), p. 257-263, 2011.

51. Li, Z. and R.H. Yoon, Thermodynamics of hydrophobic interaction between silica

surfaces coated with octadecyltrichlorosilane, Journal of colloid and interface science.

392 p. 369-375, 2013.

52. Stillinger, F.H., Water revisited, Science. 209(4455), p. 451-457, 1980.

53. Wernet, P., et al., The structure of the first coordination shell in liquid water, Science.

304(5673), p. 995-999, 2004.

54. Nilsson, A., et al., X-ray absorption spectroscopy and X-ray Raman scattering of water

and ice; an experimental view, Journal of Electron Spectroscopy and Related

Phenomena. 177(2–3), p. 99-129, 2010.

55. Mallamace, F., et al., NMR evidence of a sharp change in a measure of local order in

deeply supercooled confined water, Proceedings of the National Academy of Sciences.

105(35), p. 12725-12729, 2008.

56. Mallamace, F., The liquid water polymorphism, Proceedings of the National Academy of

Sciences. 106(36), p. 15097-15098, 2009.

57. Potoczny, Z.M., et al., Surface Tension of Bitumen. I. Temperature Dependence,

AOSTRA Journal of Research. 1 p. 107-115, 1984.

58. Israelachvili, J.N., Intermolecular and surface forces: revised third edition, p. 266,

Academic press, 2011.

59. Dai, Q. and K.H. Chung, Bitumen—sand interaction in oil sand processing, Fuel. 74(12),

p. 1858-1864, 1995.

101

60. Liu, J., et al., Bitumen–clay interactions in aqueous media studied by zeta potential

distribution measurement, Journal of colloid and interface science. 252(2), p. 409-418,

2002.

61. Ramachandran, R. and P. Somasundaran, Effect of temperature on the interfacial

properties of silicates, Colloids and Surfaces. 21(0), p. 355-369, 1986.

62. Srinivasan, N.S., L.G. Spitzer, and L.G. Hepler, J. Can. Petr. Technol. 21(4), p. 25, 1982.

63. Somasundaran, P. and R.D. Kulkarni, A new streaming potential apparatus and study of

temperature effects using it, Journal of Colloid and Interface Science. 45(3), p. 591-600,

1973.

64. McShea, J.A. and I. Callaghan, Electrokinetic potentials at the gas-aqueous interface by

spinning cylinder electrophoresis, Colloid & Polymer Science. 261(9), p. 757-766, 1983.

102

Chapter 4

The Effect of Solution Chemistry on the Stability of Wetting

Films on Bitumen

Abstract

The solution chemistry plays an important role in bitumen aeration. In this study, the effects

of solution chemistry on the thinning of wetting films between bitumen and air bubble were

studied. It was found that the thinning kinetics of wetting films decreased with increasing

immersion time. The addition of electrolyte KCl compressed the double layer and increased the

thinning kinetics. The presence of calcium and magnesium ions did not show significant effects

on the thinning kinetics when 1 mM KCl was present. In addition, the effect of approach speed of

air bubble to bitumen surface was investigated. It was found that dimples were formed at high

approaching speeds. There existed a critical film thickness, above which the wetting film thinned

faster at higher approaching speeds and below which the film thinned faster at lower approaching

speeds.

4.1 Introduction

Properties of the thin liquid films (TLFs) formed between particles, bubbles, and drops

control their interactions with each other. In Athabasca bitumen flotation in Canada, air bubbles

collide with bitumen droplets and create wetting films between them. If the TLFs are unstable,

bitumen droplets coalesce, attach to air bubbles, and float. In contrast, no flotation is possible if

103

the TLFs are stable. Thus, the stability of wetting films is of critical importance in bitumen flotation,

and is strongly dependent on the chemistry of the flotation system, which includes factors such as

temperature, pH, electrolyte concentration, the presence of fines, etc.

The effect of processing temperature on bitumen extraction has been studied extensively.1-

7 It has been reported that bitumen recovery does not significantly change in the range of 50 to

80 °C, but sharply decreases with decreasing temperatures below 35 °C (see Chapter 2).

The solution pH is another critical operating parameter, which controls the surface charges

of the particles in bitumen slurry. As early as 1920s, in the Clark’s Hot Water Extraction Process,

chemicals such as NaOH and Na2CO3 were added to the oil sands slurry to increase the solution

pH to achieve satisfactory bitumen recovery.8 It was found that there was an optimal concentration

of reagent to obtain maximum bitumen recovery.9 Sanford and Seyer10 found that the role of NaOH

addition was to derive natural surfactants by ionizing organic acids in the bitumen. However, a

subsequent study indicated that only a small fraction of NaOH was needed to produce sufficient

natural surfactants needed for bitumen flotation, and that the bulk of added NaOH reacted with

polyvalent metal carbonates, clays and sulfates contained in the oil sands.11 One study also

produced a report that increasing slurry pH favors bitumen detachment (or liberation) from sand

grains.4 Specifically, the zeta potential of the bitumen surface in an aqueous solution becomes

more negative with increasing pH.4, 12, 13 After the pH had reached 8, the zeta potential leveled off.

Moreover, the zeta potential of an asphaltene surface becomes increasingly negative as the pH is

increased.14-17 Force measurements conducted by AFM showed that the repulsive force between

bitumen-silica, asphaltene-silica, bitumen-bitumen and asphaltene-asphaltene in an aqueous

solution increased with increasing pH.14, 18-21

104

Divalent cations such as Ca2+ and Mg2+ ions are always present in a bitumen extraction

system. Calcium ions can be introduced into a bitumen extraction system through gypsum, which

is added as a process aid to form consolidated tailings in the recycling of process water from

tailings.22 The presence of Ca2+ ions could decrease the absolute values of the negative zeta

potentials of bitumen and clay surfaces due to their adsorption on these surfaces.13, 18, 23 A sizable

number of studies also confirmed that the presence of calcium and magnesium ions has a

detrimental effect on bitumen recovery.11, 18, 24, 25 However, Kasongo et al.23 found that calcium

alone appears to have a marginal effect on bitumen recovery. When both calcium and

montmorillonite were present, however, bitumen recovery decreased sharply. In support of

Kasongo’s findings, Gu et al.26 reported that calcium and fine solids had a synergic effect on the

induction time of air-bubble attachment. The researchers proposed that when calcium ions are

combined with fine solids, the adsorbed positive calcium ions on fines would interact with negative

carboxylic groups on bitumen surfaces, acting as a ‘‘binder’’ between bitumen and fine solids. A

coating of fine solids on a bitumen surface changes its wettability from hydrophobic to more

hydrophilic in nature, thereby reducing bitumen-bitumen coalescence and air bubble-bitumen

attachment.

Although earlier studies have advanced our knowledge of bitumen extraction from oil

sands, we still do not fully understand the stability of the wetting films formed between bitumen

and air bubbles. The present study was designed to fill that knowledge gap by investigating the

various factors that impact the thinning of the wetting films of water formed on bitumen.

Specifically, a bubble-against-plate (BAP) apparatus was used to measure the kinetics of the

thinning of wetting films of water formed between bitumen and bubble under different conditions.

Optical interference patters (Newton rings) were recorded using a high-speed camera and then

105

analyzed offline to calculate the film thickness, which changed with time at different radial

positions of the film. The results will be useful in elucidating the mechanisms involved in bitumen

aeration.

4.2 Experimental

4.2.1 Materials

A sample of bitumen (99 wt%) was provided by Suncor Energy, Inc. Reagent grade toluene

(>99.5% purity) was purchased from Spectrum Chemical Mfg. Corp. Anhydrous CaCl2 (99%, Alfa

Aesar) and MgCl2 (99%, Alfa Aesar), and KCl (99%, Alfa Aesar) were used to study the effect of

electrolyte on the film thinning. Diluted hydrochloric acid and NaOH were used to adjust solution

pH. Millipore ultrapure water with a resistivity of 18.2 MΩ·cm was obtained using a Direct Q3

water purification system. Silicon wafers (100 crystal planes, University Wafer) were used as the

substrates for coating bitumen. Sulfuric acid (H2SO4, 98% purity, VMR international) and

hydrogen peroxide (H2O2, 29.0–32.0% purity, Alfa Aesar) were used to clean the silicon substrates.

4.2.2 Preparation of Bitumen-coated Surfaces

Silicon wafers were first oxidized in an oxidation furnace at 1070 °C for ~40 min using

wet oxidation procedure. The oxidized silicon wafers were then cut into 12x12 mm pieces and

used as substrates for bitumen, asphaltene, and maltene. Before the coating, the silicon substrates

were put in an ultrasonic bath of water for ~1 hr and then cleaned by immersing in a Piranha

solution (3:7 by volume of H2O2/H2SO4) at 120 °C for 1 hr, followed by rinsing with an adequate

amount of ultrapure water and pure nitrogen (N2) blow-drying. The silicon substrate was placed in

a spin-coater (Laurell, WS-400-6NPP), rinsed with toluene, and then spun at 6,000 rpm for 30 s to

dry off the toluene. When the substrate was free of toluene, a small amount of (2 or 3 drops) of a

bitumen-in-toluene solution (2.5 mg/mL) was placed on the substrate and let it stand for 10 s. The

106

substrate was then spun at 3,500 rpm for 20 s before placing another drop of the bitumen solution

and continuing spinning for additional 10 s. The spinning continued for an additional 1 minute at

6,000 rpm to evaporate the solvent and obtain a smooth and uniform bitumen coating. The spin-

coated bitumen was then placed in a particulate matter-free laminar hood for ~30 min to ensure

maximum removal of toluene from the coating.

4.2.3 Measurement of Film Thinning Kinetics

In the TFPB technique described in previous chapters, a small volume of liquid is used to

measure the thickness of wetting films. Also, the experimental procedure is complex and delicate.

The objective of the present work is to study the effect of solution chemistry on the stability of

thin liquid films, for which it would be desirable to use larger volumes of liquids. Figure 4.1 shows

the bubble against plate (BAP) apparatus designed and constructed in the present work for this

purpose. In a given experiment, a solution of known ionic composition is placed in a rectangular

Figure 4.1 Schematic representation of the bubble-against-plate (BAP) apparatus used to

monitor the film thinning kinetics of wetting films.

107

Plexiglas cell with dimensions of 38x38x14 mm. The bottom plate of the cell is thick enough to

fixate a short glass tubing (~1.8 mm OD) at the center. Once a bubble of 3 mm diameter is formed

on the tip of the glass tubing by means of an air-tight syringe, a flat silicon plate (9x9x0.5 mm)

coated with bitumen is brought close to the bubble surface within a distance of 15 µm. The bubble

is then moved upward by means of the piezoelectric crystal at an approach velocity of 0.2 µm/sec.

As the film thickness is reduced, optical interference patters (Newton rings) begin to appear. The

Newton rings are recorded as a function of time by means of a high-speed camera using the same

optical system employed for the TFPB system. The recorded Newton rings are analyzed off-line

to determine the changes in film thickness (h) as a function of time (t) and radial distance (r). In

the present work, all experiments have been carried out at room temperature, i.e., 22 °C.

4.3 Results and Discussion

4.3.1 Effect of Immersion Time

Figure 4.2 shows the results of the film thickness measurements conducted using the

bubble against plate (BAP) apparatus shown in Figure 4.1. The measurements were conducted

after immersing bitumen-coated silicon plates for different periods of time, i.e., 5, 15, 30, and 60

min. As shown, the kinetics of film thinning decreased with increasing immersion time, indicating

that the bitumen surface becomes less hydrophobic at longer immersion times. The results show

also that the equilibrium film thickness (he) increased with increasing immersion time; he increased

from 118 nm to 156 nm when the immersion time was increased from 5 min to 1 h. Note, however,

that neither the thinning kinetics nor the equilibrium thickness increased further after 30 min of

immersion time, indicating that 30 min is sufficient for the bitumen film to reach an equilibrium

with water at room temperature. This finding is in agreement with the work conducted by Liu, et

al.19

108

The time-dependent behavior of bitumen may be attributed to the slow kinetics of the polar

groups of the asphaltene and resin, e.g., carboxyl, amide, thiol, and hydroxyl groups,27 migrating

through the polymeric matrix at room temperature. When a bitumen surface is exposed to air,

molecules in bitumen lie collapsed on the surface.12 Upon exposure to water, the molecules swell,

as indicated by the increase in he, and the polar groups migrate toward the bitumen/water interface,

as evidenced by the slow film thinning kinetics. The slow kinetics of the ‘swelling’ and the

migration of polar species may be due to the high viscosity of bitumen at room temperature. At

higher temperatures, the kinetics of migration are faster, and the equilibrium thickness can be

reached faster. The results presented in Figure 4.2 are consistent with the general observation that

the surface chemistry of bitumen in contact with water is time-dependent.18-20

0 10 20 30 40 50

100

200

300

5 min

15 min

30 min

1 hh

(n

m)

t (s)

Increasing time

Figure 4.2 Thinning kinetics of the wetting films formed on bitumen at different

immersion times in water at pH 8.4 to 7.5.

109

4.3.2 Effect of Electrolyte KCl

Figure 4.3 shows the effect of KCl on the thinning kinetics of the wetting films formed on

bitumen. The experiments were conducted at a natural pH of about 5.7 and room temperature (~22

°C). In the absence of KCl, the wetting film thinned slowly and reached a thickness of 153 nm in

50 s. The film did not rupture. In the presence of 1 mM KCl, the film thinned faster and ruptured

within 9 s at a critical film thickness (hcr) of 34 nm.

As shown in the foregoing section, the bitumen surface is negatively charged in water due

to the exposure of the negatively-charged polar groups on the surface, with its zeta potential being

around -60 mV.13 The air bubble surface is also negatively charged with different values of

negative zeta potential reported in the literature.28-31 Therefore there exists a repulsive electrostatic

force between bitumen and air bubble, creating a positive disjoining pressure and thus causing the

thinning rate to decrease and stop after reaching the equilibrium thickness.

0 10 20 30 40 500

100

200

300

1 mM KCl

h (

nm

)

t (s)

0 mM KCl

rupture

Figure 4.3 Effect of KCl on the thinning kinetics of the wetting films formed on

bitumen in aqueous solution of pH 8.4 to 7.5 and at 22 °C.

110

In the presence of KCl, the electrical double-layers at the bitumen/water and air/water

interfaces are compressed, causing the positive disjoining pressure to decrease. There are two

consequences of the double-layer compression: one is to increase the kinetics of film thinning, and

the other is to decrease the film thickness, both of which are clearly manifested in Figure 4.3. As

the film thickness becomes thinner, the film is subjected to a stronger hydrophobic force, which is

the major destabilizing force for wetting films, and therefore ruptures.

4.3.3 Effect of Solution pH

Figure 4.4 shows the effect of solution pH on the kinetics of film thinning in 1 mM KCl

solution. As shown, the kinetics of film thinning decreased with increasing pH. The pH

adjustments were made by adding aliquots of HCl and NaOH solutions. At pH < 5, the film

ruptured within less than 3 s, while at pH ~8, the film was stable for as long as 9 s before it finally

ruptured. These values are close to what has been reported earlier.12 In general, the film rupture or

induction time for bitumen flotation is orders of magnitude longer than observed in mineral

0 3 6 90

100

200

300

3.1

5.3-4.8

8.4-7.5

h (

nm

)

t (s)

pH

Increasing pH

Figure 4.4 Effect of solution pH on the thinning kinetics of the wetting films formed

on bitumen in a 1 mM KCl solution. The short arrows indicate the critical rupture

thicknesses (hcr) of the wetting films.

111

flotation, which has been attributed to the exposure of asphaltene on a bitumen surface. As has

already been noted, the polar groups of asphaltene include carboxyl, amide, thiol, and hydroxyl

groups, which are more readily ionized at higher pHs, giving rise to an increased electrostatic

repulsive force. At the same time, the thickness of the asphaltene layer tend to increase with

increasing pH, causing stronger steric repulsion. Both of these forces contribute to the increased

stability of the wetting films formed on bitumen and hence to the slower kinetics of film thinning

at higher pHs.

4.3.4 Effect of Calcium Ions

Figure 4.5 shows the changes in the thickness of wetting films between the bitumen surface

and air bubble in 1 mM KCl solution at different CaCl2 concentrations. When the calcium

concentration was increased from 0 to 1 mM, the kinetics of film thinning slightly increased. As

the concentration was further increased to 10 mM, however, the kinetics decreased. The initial

increase in kinetics may be attributed to the likelihood that the Ca2+ ions react with the carboxylate

0 3 6 90

100

200

300

0

0.1

1

10

h (

nm

)

t (s)

Ca2+

(mM)

Figure 4.5 Effect of Ca2+ ions on the thinning kinetics of the wetting films formed

on bitumen in a 1 mM KCl solution at pH 8.4 to 7.5.

112

groups (RCOO−) of the asphaltene and, thereby, reduce the negative surface charge and hence the

positive disjoining pressure. At 10 mM calcium concentration, it is possible that calcium hydroxyl

or carbonate species adsorb on the surface, rendering the bitumen surface hydrophilic. In general,

the critical rupture thickness decreased with increasing calcium concentration, as shown in Figure

4.5, which may be attributed to the likelihood that in the presence of Ca2+ ions the wetting films

become thinner, subjecting the film to stronger hydrophobic forces. Overall, the changes in the

film thinning kinetics and the critical rupture thickness at different calcium concentrations were

not significant compared to other factors.

4.3.5 Effect of Magnesium Ions

Figure 4.6 shows the effect of MgCl2 concentration on the thinning kinetics of the wetting

films formed on bitumen in 1 mM KCl solution at pH 8.4-7.5. As shown, the addition of Mg2+ ions

had a relatively small effect on the kinetics of film thinning but a relatively large effect on rupture

0 3 6 90

100

200

300

0

0.1

1

10

h (

nm

)

t (s)

Mg2+

(mM)

Figure 4.6 Effect of Mg2+ ions on the thinning kinetics of the wetting films formed

on bitumen in a 1 mM KCl solution at pH 8.4 to 7.5.

113

time. The rupture time decreased substantially with increasing concentration of Mg2+ ions in

solution, which may also be attributed to the decrease in the electrostatic repulsion.

Masliyah, et al.3 studied the effect of Mg2+ and Ca2+ ions on laboratory flotation, and

showed that the effect was marginal. Thus, the results obtained in the present work is consistent

with the work of Masliyah et al. A detrimental impact was observed, however, with some ores in

the presence of montmorillonite clay.

4.3.6 Effect of Approaching Speed on Film Thinning

Much of the study conducted in the present work was to study the surface chemistry

variables. However, the interaction force between bubble and bitumen should also be affected by

the hydrodynamic force. To better understand the role of the hydrodynamic force, the velocity of

the bubble approaching a flat bitumen surface was varied during the film thickness measurement

using the BAP apparatus (Figure 4.1). Figure 4.7 compares the temporal and spatial profiles of the

wetting films formed at approach velocities of 0.2 and 4.6 µm/s in a 1 mM KCl solution with

natural pH of about 5.7. At the lower approaching velocity, the film thinned slowly and ruptured

at the critical film thickness (hc) of 40 nm in 2.78 s. At the higher approaching speed, the film

-80 -40 0 40 80

0

300

600

9004.6 m/s

-50 -25 0 25 50

0

300

600

9000.2 m/s

(b)

r (m)

h (

nm

)

r (m)

h (

nm

)

0 s

(a)

3.37

0.7

0.27

0 s

2.78

1.5

0.67

Figure 4.7 Temporal and spatial profiles of the wetting films formed bitumen in a 1 mM KCl

solution at two different approach speeds: a) 0.2 and b) 4.6 m/s.

114

became flat initially and then began to form a dimple after 0.27 s. At longer thinning times, the

dimple became increasingly convex, forming a well-defined barrier rim at r = 73 µm. At 3.37 s,

the film ruptured catastrophically at the rim, with a rupture thickness of 70 nm. Similar

observations were made on hydrophobic surfaces by Pan et al.32

In Figure 4.8, the closest separation distances between bubble and bitumen have been

plotted as a function of time at the two different approach speeds employed in the present work,

i.e., 0.2 and 4.6 µm/s. Also shown in the figure is a plot of film thicknesses at the center of the

dimpled film. Initially, the dimpled film thinned faster than the non-dimpled film. After 1.7 s, the

non-dimpled film began to thin faster and ruptured before the dimpled film did. Does this mean

that less turbulence or weaker hydrodynamic force is better for bitumen aeration? It is difficult to

answer this question on the basis of a single set of experiments. In principle, dimpled films should

0 1 2 30

100

200

300

h (

nm

)

t (s)

Figure 4.8 Comparison of the thinning kinetics obtained for dimpled and non-dimpled

wetting films in a 1 mM KCl solution. Black triangles represent the minimum film

thickness of the dimpled film, i.e., at the rim; black circles represent the film thicknesses

at the center of the dimpled films; and red circles, the minimum film thicknesses of the

non-dimpled films, i.e., at the center.

115

thin more slowly because they have larger film radii than non-dimpled films, as shown in Figure

4.7. On the other hand, the film thinning kinetics should also depend on the magnitude of the

hydrodynamic force applied. Thus, there may be an optimum hydrodynamic force that can give

the fastest film thinning kinetics. Further investigation is needed to find optimum hydrodynamic

conditions for bitumen aeration.

4.4 Summary

In this study, the effects of solution chemistry on the thinning of wetting films formed

between bitumen and air bubble were studied. It was found that the kinetics of film thinning

decreased with increasing immersion time. In the presence of an electrolyte (KCl), the kinetics

increased due to double-layer compression. In the presence of KCl in the solution, the addition of

divalent cations such as Ca2+ and Mg2+ ions did not change the thinning kinetics significantly.

In addition, the effect of approach speed of air bubble to bitumen surface was investigated.

It was found that dimples were formed at high approach speeds, which is commonly observed with

other wetting and foam films. In general, film thinning rate is increased due to increased

hydrodynamic pressure; however, its effect is more significant on the rim of a film where the

curvature (or Laplace) pressure is higher than at the center of the film and hence the dimple

formation.

References

1. Long, J., Z. Xu, and J. Masliyah, On the role of temperature in oil sands processing,

Energy & fuels. 19(4), p. 1440-1446, 2005.

116

2. Long, J., et al., Effect of Operating Temperature on Water-Based Oil Sands Processing,

The Canadian Journal of Chemical Engineering. 85(5), p. 726-738, 2007.

3. Masliyah, J., et al., Understanding Water‐Based Bitumen Extraction from Athabasca

Oil Sands, The Canadian Journal of Chemical Engineering. 82(4), p. 628-654, 2004.

4. Dai, Q. and K.H. Chung, Bitumen—sand interaction in oil sand processing, Fuel. 74(12),

p. 1858-1864, 1995.

5. Schramm, L., et al., Temperature effects from the conditioning and flotation of bitumen

from oil sands in terms of oil recovery and physical properties, Journal of Canadian

Petroleum Technology. 42(8), p. 55-61, 2003.

6. Ding, X., Effects fo Divalent Ions, Illite Clays and Temperature on Bitumen Recovery, in

Department of Chemical and Materials Engineering, University of Alberta, Edmonton,

AB, 2006.

7. Zhou, Z., et al., Assessment of Bitumen Re c overy from the Athabasca Oil Sands Using a

Laboratory Denver Flotation Cell, The Canadian Journal of Chemical Engineering.

82(4), p. 696-703, 2004.

8. Clark, K. and D. Pasternack, Hot Water Seperation of Bitumen from Alberta Bituminous

Sand, Industrial & Engineering Chemistry. 24(12), p. 1410-1416, 1932.

9. Sanford, E.C., Processibility of athabasca oil sand: Interrelationship between oil sand

fine solids, process aids, mechanical energy and oil sand age after mining, The Canadian

Journal of Chemical Engineering. 61(4), p. 554-567, 1983.

10. Sanford, E.C. and S. F.A., Processability of Athabasca Tar Sand Using a Batch

Extraction Unit: The Role of NaOH, CIM Bull. 72(803), p. 164-169, 1979.

117

11. Smith, R.G. and L.L. Schramm, The influence of mineral components on the generation

of natural surfactants from Athabasca oil sands in the alkaline hot water process, Fuel

Processing Technology. 30(1), p. 1-14, 1992.

12. Yoon, R.H., D. Guzonas, and B.S. Aksoy, Role of Surface Forces in Tar Sand

Processing, in Proceeding of the 1st UBC-McGill Bi-Annual International Symposium

Vancouver, BC, Canada, 1995.

13. Liu, J., et al., Bitumen–clay interactions in aqueous media studied by zeta potential

distribution measurement, Journal of colloid and interface science. 252(2), p. 409-418,

2002.

14. Abraham, T., et al., Asphaltene-silica interactions in aqueous solutions: direct force

measurements combined with electrokinetic studies, Industrial & engineering chemistry

research. 41(9), p. 2170-2177, 2002.

15. Parra-Barraza, H., et al., The zeta potential and surface properties of asphaltenes

obtained with different crude oil/n-heptane proportions, Fuel. 82(8), p. 869-874, 2003.

16. González, G., et al., Electrokinetic characterization of asphaltenes and the asphaltenes-

resins interaction, Energy & fuels. 17(4), p. 879-886, 2003.

17. Vega, S.S., et al., The zeta potential of solid asphaltene in aqueous solutions and in 50:

50 water+ ethylene glycol (v/v) mixtures containing ionic surfactants, Journal of

Petroleum Science and Engineering. 69(3), p. 174-180, 2009.

18. Liu, J., Z. Xu, and J. Masliyah, Studies on bitumen-silica interaction in aqueous solutions

by atomic force microscopy, Langmuir. 19(9), p. 3911-3920, 2003.

118

19. Liu, J., Z. Xu, and J. Masliyah, Colloidal forces between bitumen surfaces in aqueous

solutions measured with atomic force microscope, Colloids and Surfaces A:

Physicochemical and Engineering Aspects. 260(1), p. 217-228, 2005.

20. Yoon, R.H. and Y.I. Rabinovich, Role of Asphaltene in the Processing of Tar Sand, in

Proceeding of the 3rd UBC-McGill Bi-Annual International Symposium, CIM, Montreal,

Quebec, Canada, 1999.

21. Liu, J., et al., Colloidal interactions between asphaltene surfaces in aqueous solutions,

Langmuir. 22(4), p. 1485-1492, 2006.

22. Consortium, F.T.F., A.D.o.E.O. Sands, and R. Division, Advances in oil sands tailings

research, Alberta Department of Energy, Oil sands and Research Division, 1995.

23. Kasongo, T., et al., Effect of clays and calcium ions on bitumen extraction from

athabasca oil sands using flotation, The Canadian Journal of Chemical Engineering.

78(4), p. 674-681, 2000.

24. Speight, J.G. and S.E. Moschopedis, Factors affecting bitumen recovery by the hot water

process, Fuel Processing Technology. 1(4), p. 261-268, 1978.

25. Takamura, K. and R.S. Chow, A mechanism for initiation of bitumen displacement from

oil sand, Journal of Canadian Petroleum Technology. 22(06), 1983.

26. Gu, G., et al., Effects of physical environment on induction time of air–bitumen

attachment, International Journal of Mineral Processing. 69(1), p. 235-250, 2003.

27. Lesueur, D., The colloidal structure of bitumen: Consequences on the rheology and on

the mechanisms of bitumen modification, Advances in colloid and Interface Science.

145(1), p. 42-82, 2009.

119

28. Yoon, R.-H. and J.L. Yordan, Zeta-potential measurements on microbubbles generated

using various surfactants, Journal of Colloid and Interface Science. 113(2), p. 430-438,

1986.

29. McShea, J.A. and I. Callaghan, Electrokinetic potentials at the gas-aqueous interface by

spinning cylinder electrophoresis, Colloid & Polymer Science. 261(9), p. 757-766, 1983.

30. Li, C. and P. Somasundaran, Reversal of bubble charge in multivalent inorganic salt

solutions—Effect of magnesium, Journal of Colloid and Interface Science. 146(1), p. 215-

218, 1991.

31. Creux, P., et al., Specific Cation Effects at the Hydroxide-Charged Air/Water Interface,

The Journal of Physical Chemistry C. 111(9), p. 3753-3755, 2007.

32. Pan, L., S. Jung, and R.H. Yoon, A fundamental study on the role of collector in the

kinetics of bubble–particle interaction, International Journal of Mineral Processing. 106

p. 37-41, 2012.

120

Chapter 5

Summary and Future Work

5.1 Research Revisits

Flotation is the major separation process used for the recovery of bitumen from oil sands.

The process involves two steps, bitumen liberation and aeration. The aeration step involves bubble

attachment on bitumen, followed by engulfment of the bubble by the bitumen drop. In the present

work, the kinetics of aeration (or bubble-bitumen interaction) have been studied. Most of the work

focused on the measurement of disjoining pressures of the water films of thickness h formed on

bitumen and on its hydrophobic and hydrophilic components, i.e., maltene and asphaltene. This

was accomplished using a modified thin film pressure balance (TFPB) and a new bubble against

plate (BAP) apparatus developed in the present work. Both were equipped with an optical

microscope, microinterferometry, and a high-speed video camera. The nano-scale images provided

information on the kinetics of film thinning in vertical directions, i.e., 𝜕ℎ 𝜕𝑡⁄ and spatial variation

in radial directions, i.e.,𝜕ℎ 𝜕𝑟⁄ , which are then used to determine the disjoining pressures () in

TLFs.

It is believed that the results of the present study have improved the understanding of the

complex mechanisms involved in bubble-bitumen interactions, which is very different from the

bubble-mineral and bubble-coal interactions. With improved understanding, it is now possible to

develop a bitumen flotation model and a computer simulator that can have short-term industrial

applications.

The major research conclusions and suggested future work are presented below.

121

5.2 Conclusions

It has been found that the film thinning rates increased in the order of asphaltene, bitumen,

and maltene, suggesting that the hydrophobicity of the substrates increases in the same order.

The experimental work has been carried out at temperatures ranging from 22 to 80 °C to

better understand the mechanisms involved in film thinning in the hot-, warm-, and cold-water

processing of bitumen. In general, the kinetics of film thinning increased with temperature,

indicating that the higher the temperature, the faster the film thinning and, hence, faster the aeration

kinetics. The results show also that the film thinning kinetics is sufficiently high at temperatures

as low as 35 °C. At lower temperatures, however, the film thinning and aeration kinetics are too

slow to warrant efficient flotation recovery. These results are consistent with the industrial

experience as reported in the literature.

The kinetic information on film thinning has then been used to determine the disjoining

pressure () in the wetting films. The results show that < 0 on maltene and bitumen, while >

0 on asphaltene. These results suggest that asphaltene may serve as a kinetic barrier for bubble-

bitumen interactions. In this regard, controlling asphaltene chemistry is important for the success

of bitumen flotation under various operating conditions.

The disjoining pressure data obtained in the present work have been analyzed in view of

the extended DLVO theory, which includes contributions from the electrical double-layer force,

van der Waals force, hydrophobic force, and steric force that are present in the wetting films

formed between air bubble and bitumen. It has been found that the negative disjoining pressure (

< 0) due to the hydrophobic force increases with increasing temperature, which agrees well with

the contact angles measured on thick films of bitumen. These results are also in agreement with

122

the industrial flotation practice. The disjoining pressure isotherms, established using the extended

DLVO theory, show also that steric force increases with increasing temperature.

The dynamic contact angle measurements conducted on bitumen show that, initially,

contact angles are small but subsequently increase above 90o, obviously due to the changes in the

interfacial tensions involved. It appears that the changes in interfacial tensions are brought about

by the diffusion of maltene from underneath the asphaltene layer formed on the surface of bitumen,

causing an increase in the surface hydrophobicity of bitumen. When the asphaltene layer is

saturated with maltene, an oily substance begins to coat the bubble surface. Simple thermodynamic

calculations based on the acid-base theory suggest that the asphaltene saturated with maltene,

rather than maltene, is more likely to coat the bubble surface. In effect, asphaltene is acting as a

spreading agent for maltene. Based on the data obtained in the present work, a model for the

bubble-bitumen interaction and subsequent engulfment has been proposed. The information

obtained in the present work will be useful for developing a bitumen flotation model and a

computer simulator that may be useful for industrial application near term.

5.3 Original Contribution

1. The present work represents the first measurement of disjoining pressure in the wetting

films of water formed on bitumen.

2. It has been established that the kinetics of the thinning of the wetting films formed on

bitumen and its components, i.e., asphaltene and maltene, increased with temperature in

the order of asphaltene, bitumen and maltene.

3. The disjoining pressure () in the wetting films of water deceases with increasing

temperature in the order of asphaltene, bitumen and maltene.

4. Bitumen is aerated when the disjoining pressure is negative, i.e., < 0.

123

5. The disjoining pressure of the water films formed on asphaltene is positive, i.e., > 0,

which in turn stabilizes bitumen-in-water emulsions and prevents bitumen from aeration.

6. The bitumen droplets acquire hydrophobicity by diffusion of maltene into the layer of

asphaltene exposed on the surface.

7. Air bubble on bitumen is coated by asphaltene, or the bubble is engulfed into bitumen, in

which asphaltene helps the free energy of spreading to be negative, i.e., Gs < 0. Thus,

asphaltene acts effectively as a spreading agent for maltene on bubble surface.

5.4 Future Work

5.4.1 Studies of the Film Thinning Kinetics Using a New Cell

In the present work, the modified thin film pressure balance (TFPB) technique has been

used to study the temperature effect on the thinning of wetting films of water on bitumen and its

components. There are two limitations associated with this technique due to the design of the glass

cell used to form a wetting film: 1) only a small amount of water (less than 0.1 mL) is in contact

with the surface of interest; 2) the surface under investigation has to be flat.

In industry, bitumen is in contact with large reservoirs of aqueous solutions and are

suspended in the solutions as droplets. Earlier photomicrographic studies show that rising bitumen

droplets in a continuous pilot plant were approximately spherical under good processing

conditions.1, 2 Hence, a new experimental set-up is suggested to study the thinning of wetting films

of water formed between bitumen and air bubble. It will be useful to modify the bubble-against-

plate (BAP) apparatus described in Chapter 4 by replacing the flat bitumen substrate with a

bitumen droplet. The new set-up is shown schematically in Figure 5.1. It will be useful to study

the effects of temperature, bubble size, bitumen doplet size, approach speed, and solution

chemistry (e.g., pH, Ca2+, Mg2+, clays) on the thinning of the wetting film of water on bitumen.

124

In the TFPB technique, bitumen substrate was exposed to air before contacting the water

in the cell and this is inevitable. In this way, it is possible that submicroscopic air bubbles might

form on bitumen surface since the water contact angle on bitumen surface is high (~90°). With the

new cell, one can generate a bitumen droplet in situ, i.e., in the aqueous solution without contacting

air before measurements, and ex situ, i.e., in the air and then bring the droplet into the solution,

and then compare the two sets of data to see whether the possibly adsorbed air on the bitumen

surface plays a role in the bitumen-bubble attachment. The comparison would be useful to analyze

the hydrophobic force in the bitumen aeration.

It is also suggested to do the measurements in KCl or NaCl solutions for convenient

theoretical analysis afterwards. Under a given concentration of KCl or NaCl, it is easy to calculate

the Debye length needed to determine the disjoining pressure due to the electrostatic force. Real

Figure 5.1 New experimental set-up designed for the study of the thinning of wetting film of

water on bitumen.

125

commercial processing water used for bitumen extraction in plants contains more than 20 mM

NaCl.3, 4

5.4.2 Measurement of the Length of the Asphaltene Tails in an Aqueous Solution

In Chapter 3, the theoretical analysis about the disjoining pressure isotherm based on the

extended DLVO theory was conducted by fitting the experimental data, where the length of the

adlayer (L) of polymers present in bitumen was estimated for the sake of best fitting. It is useful to

experimentally measure the adlayer length of polymers on bitumen, asphaltene, and maltene

surfaces, respectively, in an aqueous solution under different conditions (i.e., different temperature,

pH, electrolyte concentration, etc.), in order to 1) confirm experimentally the exposure of

asphaltene on bitumen surface by the comparison of the results obtained on above there surfaces;

2) use these values in the fitting process to better understand the hydrophobic force between

bitumen and air bubble.

The polymer adlayer length can be conveniently measured using dynamic light scattering

technique, also known as photo correlation spectroscopy (PCS), where the hydrodynamic diameter

of polymer-covered particles is compared with that of bare particles.5 Viscometric methods can

also be used to determine the polymer tail length.6 Small angle neutron scattering (SANS) method

can be used to determine the polymer segment density distribution.7

5.4.3 Measurement of Surface Tension of Maltene and Maltene/water Interfacial Tension at

Different Temperatures

As mentioned earlier in Chapter 2, there’s lack of data about the surface tension of maltene

and interfacial tension between maltene and water at different temperatures. It is therefore

recommended to measure those in the future work. The results would be useful for the calculation

126

of spreading coefficient for the spreading of maltene onto air bubbles and hence provide insights

on the mechanisms of the engulfment of bitumen over air bubbles.

There are a number of methods to measure surface or interfacial tension, such as Wilhelmy

plate method, pendant drop method, Du Noüy ring method, etc.

5.4.4 Thermodynamic Studies on Surface Tension Components of Bitumen and Its Components

There is also lack of data about the surface tension components, i.e., acidic, basic,

dispersion, for bitumen, maltene, and asphaltene. Determination of these components would

provide useful information to the understanding of the bitumen-air attachment and the engulfment.

5.4.5 Aeration Studies in Micron Size

In the present work, the radius of the air bubble used in the TFPB or BAP technique is

around 2 mm, which is much larger than their actual sizes in industry. It was reported that in

bitumen recovery process the mean diameter of air bubble in a pilot plant was 216 ± 140 µm, and

that of bitumen droplet was about 280 ± 140 µm.1, 2 It is preferable to do the experiments using

bubbles in the size range of that in the pilot plant. It should be noted that although the mean bubble

size fell in the mentioned range, bubbles as large as 1000 µm and as small as < 100 µm were also

observed.2 Moran et al.4 used a micropipette technique to study the interaction between bitumen

droplets and air bubbles at the 10 µm scale, roughly 10 to 40 µm in diameter.

References

1. Houlihan, R.N., Unpublished results, Syncrude Canada Ltd, Edmonton, 1976.

2. Stevenson, P., Froth Flotation of Oil Sand Bitumen, in Foam engineering: fundamentals

and applications, John Wiley & Sons, West Sussex, United Kingdom, 2012.

127

3. Allen, E.W., Process water treatment in Canada's oil sands industry: I. Target pollutants

and treatment objectives, Journal of Environmental Engineering and Science. 7(2), p. 123-

138, 2008.

4. Moran, K., J. Masliyah, and A. Yeung, Factors affecting the aeration of small bitumen

droplets, The Canadian Journal of Chemical Engineering. 78(4), p. 625-634, 2000.

5. Baker, J.A. and J.C. Berg, Investigation of the adsorption configuration of polyethylene

oxide and its copolymers with polypropylene oxide on model polystyrene latex dispersions,

Langmuir. 4(4), p. 1055-1061, 1988.

6. Gans, C., et al., Viscometric determination of the statistical segment length of wormlike

polymers, Polymer. 39(17), p. 4155-4158, 1998.

7. Fleer, G.J., et al., Polymers at interfaces, Chapman & Hall, London, 1993.